JP2015060547A - Information processing device, information processing system, information processing method, information processing program, and terminal device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an information processing device, an information processing system, an information processing method, and an information processing program that can present information matching user's preference.SOLUTION: An information processing device includes: a preference information filtering process part which filters preference information, showing preferences of an object user and other users for respective articles, with date and time information; a similar user extraction part which calculates inter-user similarities as similarities between the object user and the other users by reference to the preference information filtered by the preference information filtering process part, and extracts a similar user similar to the object user based upon the calculated inter-user similarities; and a content extraction part which extracts contents that the similar user extracted by the similar user extraction part prefers.

Description

  The present invention relates to an information processing apparatus, an information processing system, an information processing method, and an information processing program.

  Recently, in the information collection, a question answering site connected to the Internet, in which a person who is familiar with a user's question answers the question, has been actively used. In such a question answering site, since a large number of questions are posted, it may be difficult for the user to find the question that the user is looking for. Thus, a mining system that searches for and presents questions according to user preferences has been proposed.

  For example, in the technique described in Patent Literature 1, a recommendation information generation apparatus calculates a user feature vector based on history content indicating content viewed by a user, a user feature vector, and distribution target content. A content category evaluation value calculation unit for calculating a predicted value indicating a similarity to a feature vector, a proper noun type user feature vector generated based on the extracted proper nouns by extracting a set of proper nouns from history content; A proper noun extraction unit that calculates the degree of similarity with the content feature vector of the content to be distributed, calculates an updated predicted value based on the predicted value and the degree of similarity, and recommends content determined based on the calculated updated predicted value And a content prediction calculation unit that generates and distributes the information.

  In addition, the technique described in Patent Literature 2 includes a history collection unit in which a recommendation information generation device acquires status information on which a user is placed, a user history management DB that stores the acquired status information as history information, A situation similarity calculation unit for calculating the similarity between the situation indicated by the acquired situation information and the situation indicated by the history information stored in the past, and a history for rearranging the history information based on the similarity Based on the ranking section, the order of the sorted history information, a user feature vector calculation section that calculates a user feature vector representing the user's characteristics, and generates recommendation information for the user based on the calculated user feature vector And a content rearrangement unit.

JP 2010-262383 A JP 2010-128927 A

  However, even if the techniques described in Patent Literature 1 and Patent Literature 2 are used, even if the content viewed by a certain user is content that has been viewed by a plurality of similar users with similar preferences to the user, the content is old. If there is something that has become a hot topic in the past, or something that has seasonality, the content to be mined becomes widespread and presents even those that have nothing to do with the current preferences of that user There was a drawback. As described above, there is a drawback in that information suitable for the user's preference cannot be presented.

  The present invention has been made in view of the above points, and it is an object of the present invention to provide an information processing apparatus, an information processing system, an information processing method, and an information processing program capable of presenting information that meets user preferences. To do.

  (1) The present invention has been made in order to solve the above-described problems, and a first aspect of the present invention is an information processing apparatus, which has a preference for each article of a target user and other users. The preference information filtering processing unit that filters the preference information to be displayed by date and time information, and the preference information filtered by the preference information filtering processing unit, the similarity between the users that is the similarity between the target user and the other users A similar user extraction unit that extracts a similar user similar to the target user based on the calculated similarity degree between users, and a content that the similar user extracted by the similar user extraction unit likes And a content extraction unit.

  (2) Moreover, the 2nd aspect of this invention is information processing apparatus, Comprising: With reference to the preference information which shows the preference with respect to each articles | goods of a target user and another user, the said target user and said other A similar user extraction unit that calculates a similarity between users, which is a similarity with a user, and extracts a similar user similar to the target user based on the calculated similarity between the users, and the similar user extraction unit A content extraction unit that extracts content that the similar user likes, a content filtering processing unit that filters the content extracted by the content extraction unit based on date and time information, and extracts presentation content to be presented to the target user; An information processing apparatus comprising:

  (3) Moreover, the 3rd aspect of this invention is an information processing apparatus, and the preference information filtering process part which filters the preference information which shows the preference with respect to each articles | goods of an object user and another user by date information, With reference to the preference information filtered by the preference information filtering processing unit, the similarity between the users that is the similarity between the target user and the other user is calculated, and based on the calculated similarity between the users A similar user extraction unit that extracts similar users similar to the target user, a content extraction unit that extracts content that the similar user extracts that is extracted by the similar user extraction unit, and the content that is extracted by the content extraction unit The presentation content to be presented to the target user A content filtering unit, an information processing apparatus comprising: a.

  (4) Moreover, the 4th aspect of this invention is an information processing system, Comprising: The preference information filtering process part which filters the preference information which shows the preference with respect to each articles | goods of an object user and another user by date information, , Referring to the preference information filtered by the preference information filtering processing unit, calculating an inter-user similarity that is a similarity between the target user and the other user, and based on the calculated inter-user similarity An information processing system comprising: a similar user extracting unit that extracts a similar user similar to the target user; and a content extracting unit that extracts content that the similar user extracted by the similar user extracting unit likes. System.

  (5) Moreover, the 5th aspect of this invention is the information processing method of information processing apparatus, Comprising: The preference information which filters the preference information which shows the preference with respect to each articles | goods of a target user and another user by date information Referencing the preference information filtered in the filtering process and the preference information filtering process, the similarity between the users, which is the similarity between the target user and the other user, is calculated, and the calculated similarity between the users A similar user extraction process for extracting similar users similar to the target user based on the degree, and a content extraction process for extracting content preferred by the similar users extracted by the similar user extraction process. Is an information processing method.

  (6) Moreover, the 6th aspect of this invention is an information processing program of information processing apparatus, Comprising: Preference information which shows the preference with respect to each article | item of an object user and another user to the computer of information processing apparatus is shown. A preference information filtering process step that filters by date and time information, and a preference information filtered by the preference information filtering process step are referred to, and an inter-user similarity that is a similarity between the target user and the other user is calculated. A similar user extracting step for extracting similar users similar to the target user based on the calculated similarity between users, and a content extracting step for extracting content preferred by the similar users extracted by the similar user extracting step Is an information processing program.

  According to the information processing apparatus, the information processing system, the information processing method, and the information processing program of the present invention, it is possible to present information that meets the user's preference.

It is the schematic which shows an example of a structure of the information processing system which concerns on the 1st Embodiment of this invention. It is the schematic which shows an example of the image displayed on the terminal device which concerns on the 1st Embodiment of this invention. It is a schematic block diagram which shows an example of a structure of the database server which concerns on the 1st Embodiment of this invention. It is the schematic which shows an example of the question information memorize | stored in the database server which concerns on the 1st Embodiment of this invention. It is the schematic which shows an example of the user information memorize | stored in the database server which concerns on the 1st Embodiment of this invention. It is the schematic which shows an example of the classification information memorize | stored in the database server which concerns on the 1st Embodiment of this invention. It is the schematic which shows an example of the browsing information memorize | stored in the database server which concerns on the 1st Embodiment of this invention. It is the schematic which shows an example of the reply information memorize | stored in the database server which concerns on the 1st Embodiment of this invention. It is a schematic block diagram which shows an example of a structure of the terminal device which concerns on the 1st Embodiment of this invention. It is a schematic block diagram which shows an example of a structure of the web server which concerns on the 1st Embodiment of this invention. It is a hardware block diagram which shows an example of the hardware configuration of the web server which concerns on the 1st Embodiment of this invention. It is the schematic which shows an example of the question which the target user and similar user browsed concerning the 1st Embodiment of this invention. It is explanatory drawing explaining an example of the similarity of the question 3 with respect to the question 2 which concerns on the 1st Embodiment of this invention. It is the schematic which shows an example of the definition information which concerns on the 1st Embodiment of this invention. It is the schematic which shows an example of the browsing history of the object user which concerns on the 1st Embodiment of this invention. It is the schematic which shows an example of the browsing history of the similar user which concerns on the 1st Embodiment of this invention. It is the schematic which shows an example of the browsing history of the similar user by which the filtering process which concerns on the 1st Embodiment of this invention was carried out. It is the schematic which shows an example of the recommendation information which concerns on the 1st Embodiment of this invention. It is a flowchart which shows an example of the information processing in the web server which concerns on the 1st Embodiment of this invention. It is a flowchart which shows an example of the information processing in the web server which concerns on the 2nd Embodiment of this invention. It is the schematic which shows an example of the image displayed on the terminal device which concerns on the 3rd Embodiment of this invention. It is the schematic which shows an example of the definition information which concerns on the 3rd Embodiment of this invention. It is the schematic which shows an example of the definition information which concerns on the 4th Embodiment of this invention. It is a schematic block diagram which shows an example of a structure of the web server which concerns on the 6th Embodiment of this invention. It is a flowchart which shows an example of the information processing in the web server which concerns on the 6th Embodiment of this invention. It is a flowchart which shows an example of the information processing in the web server which concerns on the 7th implementation Keita of this invention.

(First embodiment)
Hereinafter, a first embodiment of the present invention will be described in detail with reference to the drawings.
FIG. 1 is a schematic diagram illustrating an example of a configuration of an information processing system 1 according to the first embodiment of the present invention.
The information processing system 1 includes a database server 100, a web server (hereinafter sometimes referred to as “information processing device”) 200, and a terminal device 300. Note that the information processing system 1 may include a plurality of terminal devices 300.

  In the information processing system 1, one or more similar users whose preferences for an article (for example, content) are similar to the target user with respect to a user who uses the terminal device 300 (hereinafter, may be referred to as “target user”). Present your favorite content. Here, the articles are not only tangible objects such as clothes, cars, books, but also contents such as music, movies, animations, television programs, video games, software, web pages, contents in web pages (for example, web pages) Intangibles such as questions above, answers on web pages, evaluations of goods or services on web pages). In the present embodiment, a case where an article is a content, more specifically, a question on a web page will be mainly described as an example. That is, the information processing system 1 presents the content in the web page preferred by one or more similar users who have similar preferences for questions on the web page to the target user.

  In addition, content includes questions on web pages (including bulletin boards, social network services, blogs, miniblogs, etc.), answers to the questions, comments, and information on articles (eg, books) (eg, product names, Detailed information of a product), a search object (for example, a web page), or any combination thereof. In the present embodiment, a case will be described as an example where the content is mainly data indicating a question on a web page, an answer to the question, or a combination thereof. In this embodiment, a similar user whose preference is similar to the target user (hereinafter also referred to as “similar user”) is, for example, a user who has a tendency to view the same question as the target user.

  Also, the preference information indicating the user's preference includes content (for example, question) browsing history, content posting history, content reply history, article (for example, book) purchase history, content poster's age, content The user's favorite gender, the user's favorite list, the user's desired list, the user's hobbies, the user's worries, and so on. In the present embodiment, a case where the preference information is mainly a browsing history of questions will be described as an example.

  The database server 100 stores, for example, question information, classification information, user information, browsing information, answer information, and resource data. The question information is information related to the question posted by each user. The classification information is information related to the classification corresponding to each content (for example, a question). User information is information about each user. The browsing information is information regarding browsing of each question by the user, for example, browsing information indicating a browsing history. The answer information is information relating to an answer posted by the user for each question, for example, answer information indicating an answer history. The resource data is data necessary for inserting display data written in a predetermined language by inserting data as materials such as questions and answers into a web page, and displaying the configured display data. The predetermined language is a markup language such as HTML (Hyper Text Markup Language). Question information, browsing information, user information, and the like are stored in the database server 100 as a table. An example of data stored in the database server 100 will be described later.

  The web server 200 has a calculation function, an access control function, and a web application function that generates display data (for example, HTML data) to be transmitted to the terminal device 300. The web server 200 extracts information requested by the information request data from the database server 100 according to the information request data received from the terminal device 300. The web server 200 generates request result data indicating the extracted information, and transmits the generated request result data to the terminal device 300.

  The terminal device 300 receives display data from the web server 200 and displays an image of a web page corresponding to the received display data. Also, the terminal device 300 transmits information request data indicating that information is requested to the web server 200, and receives request result data indicating the requested information from the web server 200 as a response to the information request data. Then, the terminal device 300 displays the request result represented by the received request result data as an image, for example.

Next, an example of an image displayed on the terminal device 300 will be described.
FIG. 2 is a schematic diagram illustrating an example of an image displayed on the terminal device 300 according to the first embodiment of the present invention.
In this example, an image showing content including a set of questions and responses concerning medical health is shown. The upper window in FIG. 2 has the title “By Category / Browsing History Category Filtering”. This window indicates that the target user can select whether to display available content by category or whether to display content according to a browsing history category (described later). The terminal device 300 detects an operation instructing the □ mark, and selects a category (for example, “internal medicine, allergy, metabo,...) Shown on the right side thereof. A check mark superimposed on the square indicates that the category shown on the right is selected. In this example, by selecting “all categories”, “internal medicine, allergy, metabolic syndrome,...”, “Pediatrics, child rearing, ...”, “surgery, neurology, ...”, “General health consultation (medicine / symptoms, etc.),” is automatically selected.

  The lower window of FIG. 2 has the title “Medical Health Consult Q & A”. In the upper left and upper right corners of this window, an input field (search window) for receiving the name of the target user “XXX” and a character string (keyword) used for the search are displayed. For example, a search button is displayed on the right side of the input field. The target user requests to search for a question including a character string input in the input field by, for example, instructing a search by operating a search button.

In the lower left part of this window, a list of three initial portions of each question viewed by the target user as “browsing history” is displayed. “Browsing history” is one piece of preference information indicating the preference of the target user.
In the lower right of this window, as the “recommended” information, the beginning part of each question is displayed in a list of six cases.

Here, the right-pointing △ mark displayed at the right end of each column of “browsing history” and “recommendation” indicates that the question and the full text of the answer to that question are displayed when an operation indicating the position is detected. Indicates that
There are two display modes for displaying “Recommendation”: a question sentence access history mode (first display mode) and a question sentence recommendation mode (second display mode). When the display mode is the first display mode, the terminal device 300 displays the questions posted by all users. On the other hand, when the display mode is the second display mode, the terminal device 300 displays a question posted by a similar user.

FIG. 3 is a schematic block diagram illustrating an example of the configuration of the database server 100 according to the first embodiment of the present invention.
The database server 100 includes a database 110, a communication unit 120, and a CPU (Central Processing Unit) 130. The database 110 and the communication unit 120 are mutually connected to the CPU 130 via a bus (bus line).

  In the database 110, for example, a table T1 (also referred to as an item table in FIG. 4) storing question information, a table T2 (also referred to as FIG. 5, user table) in which user information is stored, and a table in which classification information is stored. T3 (FIG. 6, category table) is stored. The database 110 further stores, for example, a table T4 (FIG. 7, also referred to as a browsing history table) storing browsing information and a table T5 (FIG. 8, answer history table) storing response information. The database 110 further stores resource data.

  The communication unit 120 communicates with the web server 200. The communication unit 120 receives data representing a data acquisition command (for example, an SQL (Structured Query Language) statement) from the web server 200. Then, the communication unit 120 outputs the received data acquisition command to the CPU 130. On the other hand, the communication unit 120 transmits the data acquired from the database 110 by the CPU 130 in response to the data acquisition command to the web server 200.

  CPU 130 acquires data from database 110 based on the data acquisition command received by communication unit 120. Then, the CPU 130 transmits the acquired data to the web server 200 via the communication unit 120.

FIG. 4 is a schematic diagram illustrating an example of the question information stored in the database server 100 according to the first embodiment of the present invention.
In the illustrated example, the question information is stored in the table T1. The table T1 is formed from a set in which a question ID (identifier), a user ID (questioner ID) of a user who posted a question, a question, a question posting date, a category ID, and a subcategory ID are associated with each other. The question ID is identification information for identifying each question, and is, for example, a numerical value unique to each question. The user ID of the user who posted the question is identification information for identifying the user who posted the question identified by the question ID, and is, for example, a numerical value unique to each user. The question is a character string indicating the content of the question identified by the question ID. The question posting date and time is the date and time when the user posted the content of the question. The category ID is identification information representing the classification assigned to the question, and is a numerical value unique to each classification, for example. The subcategory ID is a subcategory that is an auxiliary classification obtained by further subdividing the category (classification), and is identification information for identifying the subcategory assigned to the question. The subcategory ID is a numerical value unique to each subcategory, for example.

  In the example shown in the second row of the table T1, the content of the question identified by the question ID “1” is “Question 1”, and the user ID of the user who posted the “Question 1” is “1001”. The question posting date and time when the user posted the “question 1” is “20YY-MM-DD TQ1”. Here, “20YY” of the posting date / time represents the year, “MM” represents the month, “DD” represents the day, and “TQ1” represents the time. Further, the category ID of the category assigned to “question 1” identified by the question ID “1” and the category ID of the subcategory are “2000” and “1”, respectively.

FIG. 5 is a schematic diagram illustrating an example of user information stored in the database server 100 according to the first embodiment of the present invention.
In the illustrated example, the user information is stored in the table T2. The table T2 is formed from a set in which a user ID, a user name, sex, and age are associated. The user ID is identification information for identifying the user as described above, and is a numerical value unique to each user, for example. The user name is the name of the user identified by the user ID. The gender is the gender of the user identified by the user ID. The age is the age of the user identified by the user ID.
In the example shown in the second row of the table T2, the user name, gender, and age of the user identified by the user ID “1001” are “xxx Taro”, “male”, and “25” years old, respectively.

FIG. 6 is a schematic diagram illustrating an example of classification information stored in the database server 100 according to the first embodiment of the present invention.
In the illustrated example, the classification information is stored in the table T3. The table T3 is formed from a set in which a category ID, a category name, a subcategory ID, and a subcategory name are associated. The category ID is identification information representing a category as described above. The category name is the name of the category identified by the category ID. The subcategory ID is identification information for identifying the subcategory as described above. The subcategory name is the name of the subcategory identified by the subcategory ID.

In the example shown in the second row of the table T3, the category name that is the name of the category identified by the category ID “2000” is “internal medicine”. The category ID “2000” is associated with the subcategory ID “1”, and the subcategory name of the subcategory identified by the subcategory ID “1” is “cardiology internal medicine”.
In the example shown in the third row of the table T3, the category name that is the name of the category identified by the category ID “2100” is “surgery”. The category ID “2100” is associated with the subcategory ID “1”, and the subcategory name of the subcategory identified by the subcategory ID “1” is “orthopedic”.
In the example shown in the 4th to 6th rows of the table T3, the category name that is the name of the category identified by the category ID “2200” is “otolaryngology”. The category ID “2200” is associated with the subcategory IDs “1”, “2”, and “3”, and the subcategory names of the subcategories identified by the subcategory IDs “1”, “2”, and “3” are “ Ear, nose, and throat.

As described above, the classification information indicates information indicating a category that is a classification of each question in a certain hierarchy, and at least one of the categories is associated with information indicating a subcategory that is a classification classified in a lower hierarchy. It has been. The classification information shown in the table T3 is information indicating a two-layer classification.
Note that the classification information may be information indicating a classification of more than two hierarchies. Further, the classification information may be information indicating the classification of one layer, for example, information in which the subcategory ID and the subcategory name are omitted in the table T3.

FIG. 7 is a schematic diagram illustrating an example of browsing information stored in the database server 100 according to the first embodiment of the present invention.
In the illustrated example, the browsing information is stored in the table T4. The table T4 is formed from a set in which the question ID, the user ID (viewer ID) of the browsing user, and the browsing date are associated. The question ID is identification information for identifying a question as described above. The user ID of the user who browsed is a user ID that identifies the user who viewed the question. The browsing date and time is the date and time when the user browsed the question. When there is an answer corresponding to the question, browsing the question includes browsing the answer corresponding to the question unless otherwise specified.

  In the example shown in the second row of the table T4, the user ID of the user who browsed the question for the question ID “1” is “1001”, and the browsing date and time when the user browsed the question is “20YY-MM-DD TB1”. Is. Here, “20YY” of the viewing date / time represents the year, “MM” represents the month, “DD” represents the day, and “TB1” represents the time. In the table T4, as shown in the third row, for the question identified by the question ID “2”, a set of “1001” and “20YY-MM-DD TB2” is set as a set of the user ID and the viewing date and time. As shown in the sixth line, a pair of “1003” and “20YY-MM-DD TB5” is associated with a pair of “1004” and “20YY-MM-DD TB7” as shown in the eighth line. ing. In this way, in the browsing information, one question ID or one or a plurality of sets of browsing date and time may be associated with one question ID.

FIG. 8 is a schematic diagram illustrating an example of answer information stored in the database server 100 according to the first embodiment of the present invention.
In the illustrated example, the answer information is stored in the table T5. The table T5 is formed from a set in which the question ID, the user ID of the respondent (answerer ID), the answer, and the answer posting date / time are associated with each other. The question ID is a question ID that identifies a question to be answered. The user ID of the respondent is a user ID that identifies the user who is the respondent who posted the answer. The answer is a character string indicating the content of the answer. The answer posting date and time is the date and time when the respondent posted the answer.

In the example shown in the second row of the table T5, the user ID of the respondent who posted the answer to the question ID “1” is “1003”, the answer is “answer 1”, and the answer posting that posted the answer The date and time is “20YY-MM-DD TA1”. Here, “20YY” in the answer posting date represents the year, “MM” represents the month, “DD” represents the day, and “TA1” represents the time. Further, “answer 1” and “answer 2” are associated with the question identified by the question ID “1”, and “answer 3” is associated with the question identified by the question ID “2”. ing.
Thus, in the answer information, one or a plurality of answers may be associated with one question.

FIG. 9 is a schematic block diagram illustrating an example of the configuration of the terminal device 300 according to the first embodiment of the present invention.
The terminal device 300 includes a storage unit 310, a communication unit 320, a CPU 330, and a display unit 340. The storage unit 310, the communication unit 320, and the display unit 340 are mutually connected to the CPU 330 via a bus (bus line).
The storage unit 310 stores various programs that the CPU 330 reads and executes.

  The communication unit 320 communicates with the web server 200. For example, the communication unit 320 transmits information request data to the web server 200 according to the control of the CPU 330. Then, the communication unit 320 receives the request result data transmitted by the web server 200 as a response to the transmission of the information request data. For example, to search for a question related to a certain keyword (for example, heart disease), the user presses the search button after inputting the character string of the keyword in the input field. In this case, the information request data represents requesting a question related to the character string. The communication unit 320 receives, for example, HTML data representing a list of questions related to the character string from the web server 200 as request result data corresponding to the information request data. Then, the communication unit 320 outputs the received request result data to the CPU 330.

  The CPU 330 executes various processes by reading a program from the storage unit 310. For example, the CPU 330 causes the display unit 340 to display the question list by outputting image data representing a list of questions included in the request result data input from the communication unit 320 to the display unit 340.

FIG. 10 is a schematic block diagram showing an example of the configuration of the web server 200 according to the first embodiment of the present invention.
The web server 200 includes an input / output control unit 210 and an extraction unit 220. The input / output control unit 210 includes a response control unit 211 and a database reading unit 212. The response control unit 211 includes a character string acquisition unit 214, a display mode acquisition unit 215, a specific content acquisition unit 216, and a category acquisition unit 217. The extraction unit 220 includes a similar user extraction unit 221, a content extraction unit 222, a similar content extraction unit 223, a season / period acquisition unit 224, and a filtering processing unit 225.

  The response control unit 211 receives the information request data from the terminal device 300 and outputs the received information request data to the extraction unit 220. Further, the response control unit 211 outputs the request result data input from the database reading unit 212 to the terminal device 300.

Hereinafter, the processing of the response control unit 211 will be described in more detail.
The character string acquisition unit 214 acquires a character string input by the target user from the terminal device 300. The character string acquisition unit 214 outputs the acquired character string to the content extraction unit 222 of the extraction unit 220 described later.
The display mode acquisition unit 215 acquires display mode data representing the display mode selected by the target user from the terminal device 300. The display mode acquisition unit 215 outputs the acquired display mode data to the content extraction unit 222 of the extraction unit 220 described later.

  The specific content acquisition unit 216 acquires, for example, a specific question selected by the user from the presentation candidate questions as the content extracted by the content extraction unit 222 described later. The presentation candidate question is a question that is a candidate to be presented to the target user. For example, the specific content acquisition unit 216 outputs the acquired specific question to the similar content extraction unit 223 of the extraction unit 220 described later.

  The category acquisition unit 217 acquires category data representing the category selected by the target user from the terminal device 300. The category acquisition unit 217 outputs the acquired category data to the content extraction unit 222 of the extraction unit 220 described later. When the category selected by the target user represents a category (preference information category, for example, browsing history category) related to the target user's preference information, the category acquisition unit 217 extracts the acquired category data from the content of the extraction unit 220 described later. To the unit 222 and the similar content extraction unit 223.

  The database reading unit 212 reads resource data from the database 110 of the database server 100. The database reading unit 212 generates request result data by inserting the extraction result data input from the extraction unit 220 described later into the read resource data. Then, the database reading unit 212 outputs the generated request result data to the response control unit 211.

  The extraction unit 220 extracts content from content stored in the database 110 of the database server 100 based on a predetermined standard. For example, the extraction unit 220 determines a similar user whose preference for the article is similar to that of the target user, and extracts content (for example, a question) viewed by the similar user determined in this way. When extracting the content, the extraction unit 220 preferentially extracts the content stored in the database 110 of the database server 100, for example, the content having a larger number of browsing by similar users.

Below, the process of the extraction part 220 is demonstrated in detail.
The similar user extraction unit 221 extracts similar users with similar preferences for the target user and the article. The similar user extraction unit 221 uses, for example, the target user and other scores as scores that increase as the number of contents (for example, questions) that the target user and other users browse in common (hereinafter, the number of common browsing) increases. The similarity between users which is a similarity with a user is calculated. Specifically, the similar user extraction unit 221 calculates the degree of similarity between users using, for example, Expression (1).

  The similarity between users has a maximum value of 1 and a minimum value of 0. The similar user extraction unit 221 calculates the similarity between users for each of the users other than the target user. The similar user extraction unit 221 sets other users other than the target user included in the top N ranks (N is a predetermined integer of 1 or more, for example, 5) as the similar users, with the calculated high similarity between users. Extract. The similar user extraction unit 221 outputs the calculated degree of similarity between users and information indicating the extracted similar user to the content extraction unit 222.

The content extraction unit 222 extracts content from the database 110 of the database server 100 based on the similarity between users calculated by the similar user extraction unit 221. The content extraction unit 222 extracts, for example, content (for example, a question) viewed by a similar user extracted by the similar user extraction unit 221 from the database 110 of the database server 100, and content that the target user has already viewed from the extracted content. Is excluded. That is, the content extraction unit 222 extracts content obtained by excluding content that has already been viewed by the target user from content that has been viewed by similar users.
Thereby, the information processing system 1 can present to the target user content excluding the content preferred by the target user among the content preferred by the similar user. For this reason, the target user can find the content being searched for more quickly.

  When the same content is viewed by a plurality of similar users, for example, the content extraction unit 222 sums up the similarities between the users of the plurality of similar users for the content, and recommends the total number for the content Calculated as a score (also referred to as a recommendation score). The recommendation score is a score indicating the degree of recommendation of the content to the target user, that is, the degree of presentation prior to the target user. Then, the content extraction unit 222 extracts the content in descending order of the recommendation score, narrows down the extracted content to a predetermined number, and arranges the extracted content in descending order of the recommendation score.

  That is, the content extraction unit 222 uses, for each content viewed by a plurality of similar users extracted by the similar user extraction unit 221, the target user of the content using the similarity between users of the plurality of similar users who browsed the content. A recommendation score representing the recommendation degree is calculated. Then, the content extraction unit 222 extracts content based on the calculated recommendation score.

In addition, when the display mode data input from the display mode acquisition unit 215 represents the first display mode, the content extraction unit 222 searches for content including the character string acquired by the character string acquisition unit 214 from all the contents. Extract the searched content. On the other hand, when the display mode data input from the display mode acquisition unit 215 represents the second display mode, the content extraction unit 222 includes the character string acquired by the character string acquisition unit 214 from the content already viewed by a similar user. Search the content and extract the searched content.
The content extraction unit 222 outputs extraction result data representing the extracted content to the database reading unit 212 and the filtering processing unit 225.
Thereby, the content classified into the category selected by the target user according to the operation is displayed on the terminal device 300.

  The similar content extraction unit 223 extracts similar content similar to the specific content (for example, a question) acquired by the specific content acquisition unit 216. Specifically, for example, the similar content extraction unit 223 sets a set of users who have already browsed specific content (hereinafter also referred to as a first question) and content different from the content (hereinafter referred to as a second question). The number of users that overlap with a set of users who have already browsed (also referred to as a user overlap number) is calculated. And the similar content extraction part 223 calculates the similarity of the 2nd question with respect to a 1st question based on the calculated user duplication number, for example. For example, the similar content extraction unit 223 calculates the similarity of each content with respect to the specific content by calculating the similarity with respect to content other than the second question. Then, the similar content extraction unit 223 extracts, for example, the top M contents (M is a predetermined integer equal to or greater than 1) having a high degree of similarity of the calculated content.

In addition, the similar content extraction unit 223 further extracts content classified into the category indicated by the category data input from the category acquisition unit 217 from the extracted content. The similar content extraction unit 223 outputs extraction result data representing the extracted content to the database reading unit 212.
Thereby, the content classified into the category selected by the target user according to the operation is displayed on the terminal device 300.

The season / period acquisition unit 224 identifies, as preference information, the question ID that the target user has browsed among the browsing information stored in the database 110 of the database server 100, and extracts the question corresponding to the identified question ID from the question information. To do. Then, the season / period acquisition unit 224 extracts the question posting date / time when the extracted question was posted from the question information, and refers to the definition information described later as the season in which the question posting date / time belongs as the season information. Identify.
In addition, when there is an answer to the extracted question, the season / period acquisition unit 224 specifies the question ID for the question viewed by the target user among the browsing information stored in the database 110 of the database server 100 as preference information, An answer corresponding to the identified question ID is extracted from the answer information. Then, the season / period acquisition unit 224 extracts the answer posting date / time when the answer to the question is posted from the question information, and refers to the definition information described later as the season in which the answer posting date / time belongs as the season information. Identify.

  The filtering processing unit 225 extracts (filters) content from the content (for example, questions) input by the content extraction unit 222 and the similar content extraction unit 223 based on the season information input by the season / period acquisition unit 224. To do. Specifically, the filtering processing unit 225 determines that the question posting date and time when the question posting date and time and the answer posting date and time have been posted during the season indicated by the season information among the plurality of input contents. The questions posted on the day and the answers posted on the date of posting the answer are extracted, and the questions and answers other than the extracted questions and answers are excluded.

As described above, the filtering processing unit 225 is posted on the posting date when the posting date of the content among the plurality of input contents indicates that the posting was made during the season indicated by the season information. Extract content and exclude content other than the extracted content.
Then, the filtering processing unit 225 calculates the above-described recommendation score (recommendation score) for each of the extracted contents, and arranges a predetermined number of questions in descending order of the calculated recommendation score.
The filtering processing unit 225 outputs the arranged contents to the database reading unit 212 as extraction result data.

FIG. 11 is a hardware configuration diagram illustrating an example of a hardware configuration of the web server 200 according to the first embodiment of the present invention.
The web server 200 includes a CPU 201, a storage medium 202, a drive unit 203, an input unit 204, an output unit 205, a ROM 206 (Read Only Memory), a RAM 207 (Random Access Memory), and an auxiliary storage. Unit 208 and communication unit 209.
The CPU 201, the drive unit 203, the input unit 204, the output unit 205, the ROM 206, the RAM 207, the auxiliary storage unit 208, and the communication unit 209 are connected to each other via a bus (bus line).

  The CPU 201 reads out programs and various data and controls the web server 200. The storage medium 202 is a portable storage medium such as a magneto-optical disk, a flexible disk, or a flash memory, and stores, for example, various data. The drive unit 203 is, for example, a reading device or a reading / writing device for the storage medium 202. The input unit 204 is an input device such as a mouse or a keyboard. The output unit 205 is an output device such as a display unit or a speaker. The ROM 206 is a storage medium that stores a program, for example. The RAM 207 is a storage medium that temporarily stores various data and programs, for example. The auxiliary storage unit 208 is a storage medium such as a hard disk drive or a flash memory, and stores, for example, various data. The communication unit 209 has a communication interface and is connected to the network NW by wire or wireless.

  The program read by the CPU 201 is an example stored in the ROM 206, but may be stored in the storage medium 202 or the auxiliary storage unit 208, or the program downloaded from the network NW may be stored in the storage medium 202 or the auxiliary storage unit. 208 may be stored. The various data read by the CPU 201 is an example stored in the storage medium 202 or the auxiliary storage unit 208, but may be stored in the ROM 206 or may be various data downloaded from the network NW.

Next, a recommendation score calculation process in the content extraction unit 222 will be described with reference to FIG.
FIG. 12 is a schematic diagram illustrating an example of a question viewed by a target user and a similar user according to the first embodiment of the present invention.
FIG. 12A shows a table T41 showing a list of questions viewed by each user. The table T41 has item columns for user A, user B, user C, and user D. Of user A, user B, user C, and user D, user A is the target user. User B, user C, and user D are all similar users.

Under this condition, for example, the content extraction unit 222 excludes the question that the user A (target user) has already viewed from the questions that the user B, user C, and user D have already viewed. Specifically, the “question A1” viewed by the user B is also viewed by the user A (target user), so the “question A1” viewed by the user B is excluded. Similarly, “question A3” viewed by user C is excluded.
In the table T42 shown in FIG. 12B, among the questions in the table T41, a cross is given on the question excluded by the content extraction unit 222.

Here, since the user B, the user C, and the user D are browsing the “question B1”, the content extraction unit 222 determines the similarity between the user A and the user B _RAB and the user A and the user C of the user C. A value obtained by adding the inter-similarity R _AC and the inter-user similarity R _AD between the users A and D is calculated as a recommended score S _B1 (= R _AB + R _AC + R _AD ) for the “question B1”. In addition, since the user B and the user C are browsing the “question B3”, the content extraction unit 222 includes the user-to-user similarity R _AB between the user A and the user B and the user-to-user similarity R between the user A and the user C. A value obtained by adding _AC is calculated as a recommendation score S _B3 (= R _AB + R _AC ) for “question B3”.

Similarly, since the user C and the user D are browsing the “question C3”, the content extraction unit 222 performs the similarity between the users A and C between the users _RAC and the similarity between the users A and D between the users. A value obtained by adding R _AD is calculated as a recommendation score S _C3 (= R _AC + R _AD ) for “question C3”. In addition, since the user D is browsing the “question D3”, the content extraction unit 222 sets the similarity score R _AD between the users A and D to the recommended score S _D3 (= R _AD for the “question D3”) _. ).

When the recommended score S _B3 (= R _AB + R _AC ) for “Question B3” is larger than the recommended score S _C3 (= R _AC + R _AD ) for “Question C3” (S _B3 > S _C3 ), the content extraction unit 222 Are arranged in the order of “question B1”, “question B3”, “question C3”, and “question D3”. Thereby, the terminal device 300 displays “Question B1”, “Question B3”, “Question C3”, and “Question D3” in that order.
On the other hand, when the recommended score S _C3 (= R _AC + R _AD ) for “question C3” is smaller than the recommended score S _D3 (= R _AD ) for “question D3” (S _D3 <S _C3 ), the content extraction unit 222 Are arranged in the order of “Question B1”, “Question B3”, “Question D3”, and “Question C3”. Thereby, the terminal device 300 displays “Question B1”, “Question B3”, “Question D3”, and “Question C3” in that order.

Thus, the content extraction unit 222 increases the recommendation score as the number of similar users viewed among the content viewed by similar users increases. Further, the content extraction unit 222 increases the recommendation score for the content browsed by similar users as the sum of the similarities between the similar user who browsed the content and the target user increases. Thereby, among the contents browsed by similar users, it is possible to preferentially present the contents viewed by many similar users having high inter-user similarity with the target user. As a result, the information processing system 1 can increase the probability of presenting the content that the target user is looking for, so that the target user can quickly and easily find the content that the target user is looking for. As a result, it is possible to quickly and easily know what the target user wants to know from the presented content, so that user satisfaction can be improved.
Moreover, since the information processing system 1 presents content excluding content that has already been browsed by the target user, the target user can find the content that the user is looking for earlier.

Next, processing performed by the similar content extraction unit 223 will be described with reference to FIG.
FIG. 13 is an explanatory diagram illustrating an example of the degree of similarity of the question 3 with respect to the question 2 according to the first embodiment of the present invention.
In FIG. 13, the line segment connecting the user name (for example, user A) and the question name (for example, “question 1”) is viewed by the user shown at one end of the question shown at the other end. Represents that. Specifically, “Question 1” is browsed by the user A. “Question 2” is viewed by user A, user B, and user C. “Question 3” is browsed by user C and user D. “Question 4” is browsed by user B. “Question 5” is browsed by user A. In FIG. 13, an example of the similarity (content similarity) of each question with respect to “question 2” is shown immediately below the question name.

In the example of the figure, the similar content extraction unit 223, for example, according to the calculation formula of the similarity V, V = N ₃ × 2 / (N ₁ + N ₂ ), the similarity V of “question 3” with respect to “question 2” Is calculated as 0.8 (= 2 × 2 / (3 + 2)). Here, N ₁ is the number of users who viewed the question with respect to “Question 2” (in this example, 3). N ₂ is the number of users who viewed the question with respect to “Question 3” (in this example, 2). N ₃ is the overlapping number of users of “Question 2” and “Question 3” (2 in this example). Here, the similarity V means that the larger the value, the higher the similarity between questions. Thereby, the similar content extraction part 223 can make the similarity between questions high, so that there are many users who browsed repeatedly between questions.

Next, processing performed by the content extraction unit 222, the season / period acquisition unit 224, and the filtering processing unit 225 will be described with reference to FIGS.
FIG. 14 is a schematic diagram illustrating an example of definition information according to the first embodiment of the present invention.
The table TT5 is an example of definition information. Table TT5 is No. , Season and period items. No. Is an identification number assigned to each season and period. The seasons are, for example, the four seasons of “spring”, “summer”, “autumn”, and “winter”. The period represents a period corresponding to the season. For example, in the case of “spring”, the period “March 1 to June 30” corresponds.

FIG. 15 is a schematic diagram illustrating an example of a viewing history of the target user according to the first embodiment of the present invention.
The table T6 is an example of a browsing history of the user A who is the target user, for example. Table T6 is No. , Question ID and question posting date and time. No. Represents the order when the user A browses the question corresponding to the question ID. For example, in the example of the second row, it represents that the user A browsed the question corresponding to the question ID “101”, and the posting date and time when the question was posted is the question posting date and time “2013-03-03 20:00”. It represents that. In this example, the posting date / time when the question corresponding to the question ID “102” is posted is the question posting date / time “2013-04-01 21:00”, and the posting date / time when the question corresponding to the question ID “103” is posted. Indicates that the question posting date is “2013-05-05 22:00”.
That is, any of the questions viewed by the user A corresponds to the period “March 1 to June 30” whose question posting date and time is shown in the table TT5, and the period “March 1 to June 30” is included. Corresponds to the season "spring".

FIG. 16 is a schematic diagram illustrating an example of browsing history of similar users according to the first embodiment of the present invention.
Table T7 is an example of a browsing history of user B who is a similar user. The table T8 is an example of a browsing history of the user C who is a similar user. The table T9 is an example of a browsing history of the user D who is a similar user.
Each of the tables T7, T8, T9 is No. , Question ID and question posting date and time. Since each item string is the same as that in the table T6, description thereof is omitted.

  The example of the second row of the table T7 indicates that the user B browsed the question corresponding to the question ID “101”, and the posting date and time when the question was posted is the question posting date and time “2013-03-03 20:00”. ". In other rows such as the third row of the table T7, it is indicated that the user B has browsed the question corresponding to the question ID, and the posting date / time when the question is posted is the question posting date / time. Yes.

  The example of the second row of the table T8 indicates that the user C has browsed the question corresponding to the question ID “102”, and the posting date and time when the question is posted is the question posting date and time “2013-04-01 21”. : 00 ”. In other rows such as the third row of the table T8, it indicates that the user C has browsed the question corresponding to the question ID, and that the posting date / time when the question is posted is the question posting date / time. Yes.

  The example of the second row of the table T9 indicates that the user D has browsed the question corresponding to the question ID “103”, and the posting date and time when the question is posted is the question posting date and time “2013-05-05 22”. : 00 ”. Also, in other rows such as the third row of the table T9, it indicates that the user D has browsed the question corresponding to the question ID, and that the posting date / time when the question is posted is the question posting date / time. Yes.

  Thus, as shown in the tables T7, T8, and T9, the browsing history of the similar users is associated with the identification information of the content viewed by each similar user and the date / time information such as the posting date / time when the content is posted. ing.

FIG. 17 is a schematic diagram illustrating an example of browsing history of similar users subjected to filtering processing according to the first embodiment of the present invention.
The table T10 is an example of a browsing history that has been filtered by the user B who is a similar user. The table T11 is an example of a browsing history that has been filtered by the user C who is a similar user. The table T12 is an example of a browsing history that has been filtered by the user D who is a similar user.
Each of the tables T10, T11, T12 is No. , Question ID and question posting date and time. Since each item string is the same as the tables T7, T8, and T9, description thereof is omitted.

The season / period acquisition unit 224 and the filtering processing unit 225 are based on the browsing history of the user A who is the target user, the browsing histories of the users B, C, and D which are similar users, and the definition information. 222, filtering processing is performed on the plurality of contents input by the similar content extraction unit 223, and the contents to be presented to the target user are extracted.
Specifically, the season / period acquisition unit 224 extracts the question posting date and time of each question viewed by the user A from the browsing history (table T6) of the user A, and the extracted question posting date and time is defined information (table TT5). ) And the season (for example, spring) associated with the period corresponding to the question posting date and time is extracted.

  And the filtering process part 225 extracts the content corresponding to the extracted season from the browsing history (table T7, T8, T9) of a similar user. The table T10 includes a period “3” of the definition information (table TT5) corresponding to the season “spring” extracted from the user B browsing history (table T6) by the filtering processing unit 225 from the user B browsing history (table T7). It is generated by extracting a question having a question posting date and time corresponding to “Month 1st to June 30th”.

  Similarly, the filtering processing unit 225 selects a question corresponding to the period of the definition information (table TT5) corresponding to the season extracted from the browsing history (table T6) of the user A from the browsing history (table T8) of the user C. The history information is extracted as in the table T11 by extracting the question of the posting date. Moreover, the filtering process part 225 is the question posting date and time corresponding to the period of the definition information (table TT5) corresponding to the season extracted from the browsing history (table T6) of the user A from the browsing history (table T9) of the user D. The history information is extracted as in the table T12 by extracting the questions.

FIG. 18 is a schematic diagram illustrating an example of recommendation information according to the first embodiment of the present invention.
The table T13 is an example of recommendation information presented to the target user. Table T13 is No. , Question ID and question posting date and time. Since each item string is the same as the tables T10, T11, and T12, description thereof is omitted.
The filtering processing unit 225 generates recommendation information by rearranging the history extracted from the browsing history of each similar user as in the tables T10, T11, and T12, for example, in descending order of the recommendation score. Specifically, the filtering processing unit 225 excludes questions that have already been browsed by the target user from the browsing history of each extracted similar user. Then, the filtering processing unit 225 calculates a recommendation score for each browsing history question other than the excluded questions, sorts the browsing history of similar users in descending order of the calculated recommendation score, and stores the table T13. Generate recommendations like this.

  In the present embodiment, the filtering processing unit 225 sorts the browsing histories extracted from the browsing histories of the similar users as in the tables T10, T11, and T12, for example, by rearranging them in descending order of the recommended score. The recommendation information may be generated without excluding the question viewed by the user.

In the present embodiment, the filtering processing unit 225 rearranges the questions in the order from the most recent posting date among the browsing histories extracted from the browsing histories of the similar users as in the tables T10, T11, and T12. Recommendation information may be generated without excluding questions that have already been viewed by the target user.
Further, the filtering processing unit 225 sorts the browsing histories extracted from the browsing histories of the similar users as in the tables T10, T11, and T12 in order from the latest posting date of the question, and has already been viewed by the target user. The recommended information may be generated by excluding the asked questions.

  In the present embodiment, the filtering processing unit 225 generates the recommendation information by rearranging the recommendation scores in descending order, but the browsing history extracted from the browsing history of each similar user as in the tables T10, T11, and T12. Among them, in the order of similar users having a large number of extracted browsing histories, such as a browsing history of similar users having a large number of extracted browsing histories, a browsing history of similar users having the second largest extracted browsing history, The recommended information may be generated by extracting a predetermined number of browsing histories from the browsing histories of similar users. At this time, the predetermined number is different for each stepwise similar user according to the order of similar users with many extracted browsing histories, for example, similar user B is 3, similar user C is 2, similar user D May be 1 or the same number for all similar users. In this case, in a similar user, when the number of browsing histories extracted is equal to or less than a predetermined number, the filtering processing unit 225 may extract all the browsing histories and generate recommendation information.

  In the present embodiment, the filtering processing unit 225 has the browsing history most recently viewed by each similar user among the browsing histories extracted from the browsing history of each similar user as in the tables T10, T11, and T12, for example, A predetermined number of browsing histories are extracted in descending order from the value in the item string of each identification information (No.) in the tables T10, T11, T12, and the extracted browsing histories are shown in the tables T10, T11, T12. Alternatively, the recommendation history may be generated by rearranging the browsing histories in the order of similar users having a large number of browsing histories extracted from the browsing histories of the similar users.

Next, an example of information processing performed by the web server 200 according to the present embodiment will be described.
FIG. 19 is a flowchart showing an example of information processing in the web server 200 according to the first embodiment of the present invention.
In step ST101, the similar user extraction unit 221 calculates the similarity between users between the target user and each of the other users. Then, it progresses to step ST102.
In step ST102, the similar user extraction unit 221 extracts the top N users having the similarities between users as similar users. Then, it progresses to step ST103.
In step ST103, the content extraction unit 222 extracts all questions viewed by similar users. Then, it progresses to step ST104.
In step ST104, the content extraction unit 222 excludes questions already viewed by the target user from all the questions browsed by the similar users extracted in step ST103. Then, it progresses to step ST105.

In step ST105, the season / period acquisition unit 224 acquires the season information related to the question viewed by the target user based on the question posting date and the definition information of the question viewed by the target user. Then, it progresses to step ST106.
In step ST106, the filtering processing unit 225 excludes questions posted in seasons other than the season information acquired in step ST105 from the questions extracted in step ST103. Then, it progresses to step ST107.
In step ST107, the filtering processing unit 225 calculates a recommendation score (recommendation score) for each of the questions left out in step ST106. Then, it progresses to step ST108.
In step ST108, the filtering processing unit 225 extracts a predetermined number of questions in the descending order of recommendation score (recommendation score), and causes the terminal device 300 to present the extracted questions. Thereafter, the processing according to FIG. 19 ends.

  As described above, according to the present embodiment, the information processing apparatus (web server 200) refers to the preference information indicating the preference for each article of the target user and other users, and the target user and other users. A similar user extraction unit 221 that calculates a similarity between users, which is a similarity between the two, and extracts a similar user similar to the target user based on the calculated similarity between users, and a similar user extracted by the similar user extraction unit 221 Content extracting unit 222 that extracts content that the user likes, and a content filtering processing unit (filtering processing unit 225) that filters the content extracted by content extracting unit 222 with date and time information and extracts presentation content to be presented to the target user And comprising.

  As a result, the information processing apparatus (web server 200) can extract content that has recently become a topic or seasonal content from the posting time of content viewed by a plurality of similar users who have similar preferences to a certain user. Information that matches the current preference of the certain user can be presented.

(Second Embodiment)
Hereinafter, a second embodiment of the present invention will be described in detail with reference to the drawings.
Since the information processing system 1a according to the present embodiment is the same as the information processing system 1 according to the first embodiment, description thereof is omitted. Moreover, since the database server 100, the web server 200, and the terminal device 300 which concern on this embodiment are the same as the database server 100, the web server 200, and the terminal device 300 which concern on 1st Embodiment, description is abbreviate | omitted.

FIG. 20 is a flowchart illustrating an example of information processing in the web server 200 according to the second embodiment of the present invention.
In step ST201, the similar user extraction unit 221 calculates the similarity between users between the target user and each of the other users. Then, it progresses to step ST202.
In step ST202, the similar user extracting unit 221 extracts the top N users having the similarities between users as similar users. Then, it progresses to step ST203.
In step ST203, the content extraction unit 222 extracts all questions viewed by similar users. Then, it progresses to step ST204.
In step ST204, the content extraction unit 222 excludes questions already viewed by the target user from all the questions browsed by the similar users extracted in step ST203. Then, it progresses to step ST205.

In step ST205, the filtering processing unit 225 calculates a recommendation score (recommendation score) for each of the questions left in step ST206. Then, it progresses to step ST206.
In step ST206, the filtering processing unit 225 extracts a predetermined number of questions in descending order of recommendation score (recommendation score). Then, it progresses to step ST207.
In step ST207, the season / period acquisition unit 224 acquires the season information related to the question viewed by the target user based on the question posting date and definition information of the question viewed by the target user. Then, it progresses to step ST208.
In step ST208, the filtering processing unit 225 excludes questions posted in seasons other than the season information acquired in step ST207 from the questions extracted in step ST103, and causes the terminal device 300 to present the remaining questions. Thereafter, the information processing according to FIG. 20 ends.

As described above, according to the present embodiment, the information processing apparatus (web server 200) refers to the preference information indicating the preference for each article of the target user and other users, and the target user and other users. A similar user extraction unit 221 that calculates a similarity between users, which is a similarity between the two, and extracts a similar user similar to the target user based on the calculated similarity between users, and a similar user extracted by the similar user extraction unit 221 Content extracting unit 222 that extracts content that the user likes, and a content filtering processing unit (filtering processing unit 225) that filters the content extracted by content extracting unit 222 with date and time information and extracts presentation content to be presented to the target user And comprising.
Thereby, the effect similar to 1st Embodiment can be acquired.

(Third embodiment)
Hereinafter, the third embodiment of the present invention will be described in detail with reference to the drawings.
Since the information processing system 1b according to the present embodiment is the same as the information processing system 1 according to the first embodiment, description thereof is omitted. Further, the database server 100 and the terminal device 300 according to the present embodiment are the same as the database server 100 and the terminal device 300 according to the first embodiment, and thus description thereof is omitted.

FIG. 21 is a schematic diagram illustrating an example of an image displayed on the terminal device 300 according to the third embodiment of the present invention.
The upper window in FIG. 21 is a recommendation switching window having the title “recommendation switching”. In this window, the user can select date and time information such as whether to extract recommended information by "period", whether to extract by "keyword", or whether to extract by "season". Indicates. In this example, the terminal device 300 detects, for example, that radio buttons are arranged corresponding to “period”, “keyword”, and “season” and are selected by the radio buttons, and are displayed on the right side thereof. The recommended information is extracted based on the seasonal information, for example, the keyword information corresponding to “keyword”.

  The lower window of FIG. 21 has the title “Medical Health Consultation Q & A”. In the upper left and upper right corners of this window, the name of the target user “XXX” and an input field (search window) for receiving a character string used for the search are displayed. For example, a search button is displayed on the right side of the input field. The target user requests to search for a question including a character string input in the input field by, for example, instructing a search by operating a search button.

In the lower left part of this window, a list of three initial portions of each question viewed by the target user as “browsing history” is displayed. “Browsing history” is one piece of preference information indicating the preference of the target user.
In the lower right of this window, as the “recommendation” information, a list of six beginning parts of each question including a keyword to be described later is displayed.

Here, the right-pointing △ mark displayed at the right end of each column of “browsing history” and “recommendation” indicates that the question and the full text of the answer to that question are displayed when an operation indicating the position is detected. Indicates that
There are two display modes for displaying “Recommendation”: a question sentence access history mode (first display mode) and a question sentence recommendation mode (second display mode). When the display mode is the first display mode, the terminal device 300 displays the questions posted by all users. On the other hand, when the display mode is the second display mode, the terminal device 300 displays only the questions posted by similar users.

  The configuration of the web server 200a according to the third embodiment of the present invention will be described with the aid of the configuration of the web server 200 according to the first embodiment shown in FIG. In this embodiment, the web server 200 in FIG. 10 is replaced with the web server 200a, the extraction unit 220 is replaced with the extraction unit 220a, the season / period acquisition unit 224 is replaced with the season / period acquisition unit 224a, and the filtering processing unit 225 is replaced. Is read as the filtering processing unit 225a.

  The web server 200a includes an input / output control unit 210 and an extraction unit 220a. The input / output control unit 210 includes a response control unit 211 and a database reading unit 212. The response control unit 211a includes a character string acquisition unit 214, a display mode acquisition unit 215, a specific content acquisition unit 216, and a category acquisition unit 217. The extraction unit 220a includes a similar user extraction unit 221, a content extraction unit 222, a similar content extraction unit 223, a season / period acquisition unit 224a, and a filtering processing unit 225a.

  When the web server 200a according to the present embodiment and the web server 200 according to the first embodiment are compared, the processing of the extraction unit 220a is different. Since the other configuration is the same as that of the first embodiment, the same reference numerals are given and description thereof is omitted, and different portions in the present embodiment will be mainly described.

The season / period acquisition unit 224a has the same function as the season / period acquisition unit 224 (FIG. 10), but differs from the season / period acquisition unit 224 in the following points.
The season / period acquisition unit 224a (keyword extraction unit) extracts the question posting date and time when the extracted question is posted from the question information, refers to the definition information described later, and determines the season of the period to which the question posting date and time belongs. Acquired as seasonal information. Then, the season / period acquisition unit 224a acquires keyword information representing a keyword corresponding to the acquired season information with reference to the definition information. Then, the season / period acquisition unit 224a outputs the acquired season information and keyword information to the filtering processing unit 225a.

  When there is an answer to the extracted question, the season / period acquisition unit 224a extracts the answer posting date / time when the answer to the question was posted from the question information, and the answer posting date / time is determined by referring to the definition information described later. The season of the period to which it belongs is acquired as seasonal information. Then, the season / period acquisition unit 224a acquires keyword information representing a keyword corresponding to the acquired season information with reference to the definition information. Then, the season / period acquisition unit 224a outputs the acquired season information and keyword information to the filtering processing unit 225a.

The filtering processing unit 225a has the same function as the filtering processing unit 225 (FIG. 10), but differs from the filtering processing unit 225 in the following points.
Based on the season information input from the season / period acquisition unit 224a, the filtering processing unit 225a extracts, as the first content, content for which the posted date of the extracted content corresponds to the season information. Then, based on the keyword information input from the season / period acquisition unit 224a, the filtering processing unit 225a searches the extracted first content for content including the keyword represented by the keyword information, and narrows down the content range.
The filtering processing unit 225a outputs the searched content to the database reading unit 212 as extraction result data.

FIG. 22 is a schematic diagram illustrating an example of definition information according to the third embodiment of the present invention.
The table T14 is an example of definition information. Table T14 is No. , Season and period keyword columns. No. Is an identification number assigned to each season. The seasons are, for example, the four seasons of “spring”, “summer”, “autumn”, and “winter”. The period represents a period corresponding to the season. For example, in the case of “spring”, the period “March 1 to June 30” corresponds. The keyword is a keyword (also referred to as a trend) that is characteristic in each season, for example, a keyword that represents an epidemic or symptom for each season.

  As described above, according to the present embodiment, the information processing apparatus (web server 200a) refers to the preference information indicating the preference of the articles of the target user and other users, and the target user and other users. A similar user extraction unit 221 that calculates a similarity between users, which is a similarity between the two, and extracts a similar user similar to the target user based on the calculated similarity between users, and a similar user extracted by the similar user extraction unit 221 Content extracting unit 222 that extracts content that the user likes, and a content filtering processing unit (filtering processing unit 225a) that filters the content extracted by content extracting unit 222 with date and time information and extracts presentation content to be presented to the target user And comprising.

  As a result, the information processing apparatus (web server 200a) corresponds to the date / time information from recently browsed content or seasonal content from content viewed by a plurality of similar users who have similar preferences to a certain user. It is possible to extract information that matches the current preference of the user.

(Fourth embodiment)
Hereinafter, a fourth embodiment of the present invention will be described in detail with reference to the drawings.
Since the information processing system 1c according to the present embodiment is the same as the information processing system 1 according to the first embodiment, description thereof is omitted. Further, the database server 100 and the terminal device 300 according to the present embodiment are the same as the database server 100 and the terminal device 300 according to the first embodiment, and thus description thereof is omitted.

  The configuration of the web server 200b according to the fourth embodiment of the present invention will be described with the aid of the configuration of the web server 200 according to the first embodiment shown in FIG. In this embodiment, the web server 200 in FIG. 10 is replaced with the web server 200b, the extraction unit 220 is replaced with the extraction unit 220b, the season / period acquisition unit 224 is replaced with the season / period acquisition unit 224b, and the filtering processing unit 225 is replaced. Is read as the filtering processing unit 225b.

  The web server 200b includes an input / output control unit 210 and an extraction unit 220b. The input / output control unit 210 includes a response control unit 211 and a database reading unit 212. The response control unit 211b includes a character string acquisition unit 214, a display mode acquisition unit 215, a specific content acquisition unit 216, and a category acquisition unit 217. The extraction unit 220b includes a similar user extraction unit 221, a content extraction unit 222, a similar content extraction unit 223, a season / period acquisition unit 224b, and a filtering processing unit 225b.

  When the web server 200b according to the present embodiment and the web server 200 according to the first embodiment are compared, the processing of the extraction unit 220b is different. Since the other configuration is the same as that of the first embodiment, the same reference numerals are given and description thereof is omitted, and different portions in the present embodiment will be mainly described.

The season / period acquisition unit 224b has the same function as the season / period acquisition unit 224 (FIG. 10), but differs from the season / period acquisition unit 224 in the following points.
The season / period acquisition unit 224b extracts, from the question information, the question posting date and time when the extracted question is posted, and acquires the month of the period to which the question posting date and time belongs as period information with reference to definition information described later. . Then, the season / period acquisition unit 224b outputs the acquired period information to the filtering processing unit 225b.

  When there is an answer to the extracted question, the season / period acquisition unit 224b extracts the answer posting date / time when the answer to the question was posted from the question information, and the answer posting date / time is determined by referring to the definition information described later. The month of the period to which it belongs is acquired as period information. Then, the season / period acquisition unit 224b outputs the acquired period information to the filtering processing unit 225b.

The filtering processing unit 225b has the same function as the filtering processing unit 225 (FIG. 10), but differs from the filtering processing unit 225 in the following points.
Based on the period information input from the season / period acquisition unit 224b, the filtering processing unit 225b extracts the content corresponding to the month in which the posted date of the extracted content represents the period information.

FIG. 23 is a schematic diagram illustrating an example of definition information according to the fourth embodiment of the present invention.
The table T15 is an example of definition information. Table T15 is No. , Month, and period keyword items. No. Is an identification number assigned to each month. Moreover, the month represents the month corresponding to each period from January to December, for example. The period represents a period corresponding to each month. For example, in the case of January, the period “January 1 to January 31” corresponds.

  As described above, according to the present embodiment, the information processing apparatus (web server 200b) refers to the preference information indicating the preference for each article of the target user and other users, and the target user and the other users. A similar user extraction unit 221 that calculates a similarity between users, which is a similarity between the two, and extracts a similar user similar to the target user based on the calculated similarity between users, and a similar user extracted by the similar user extraction unit 221 Content extracting unit 222 that extracts content that the user likes, and a content filtering processing unit (filtering processing unit 225b) that filters the content extracted by the content extracting unit 222 with date and time information and extracts the presented content to be presented to the target user And comprising.

  As a result, the information processing apparatus (web server 200b) associates content that has recently become a topic or content with seasonality from content browsed by a plurality of similar users who have similar preferences with a user. Can be extracted every month, and information suitable for the current preference of the user can be presented.

(Fifth embodiment)
Hereinafter, a fifth embodiment of the present invention will be described in detail with reference to the drawings.
Since the information processing system 1d according to the present embodiment is the same as the information processing system 1 according to the first embodiment, description thereof is omitted. Further, the database server 100 and the terminal device 300 according to the present embodiment are the same as the database server 100 and the terminal device 300 according to the first embodiment, and thus description thereof is omitted.
The configuration of the web server 200c according to the fifth embodiment of the present invention will be described with the aid of the configuration of the web server 200 according to the first embodiment according to FIG. In this embodiment, the web server 200 in FIG. 10 is replaced with the web server 200c, the extraction unit 220 is replaced with the extraction unit 220c, the season / period acquisition unit 224 is replaced with the season / period acquisition unit 224c, and the filtering processing unit 225 is replaced. Is read as the filtering processing unit 225c.

  The web server 200c includes an input / output control unit 210 and an extraction unit 220c. The input / output control unit 210 includes a response control unit 211 and a database reading unit 212. The response control unit 211a includes a character string acquisition unit 214, a display mode acquisition unit 215, a specific content acquisition unit 216, and a category acquisition unit 217. The extraction unit 220c includes a similar user extraction unit 221, a content extraction unit 222, a similar content extraction unit 223, a season / period acquisition unit 224c, and a filtering processing unit 225c.

  When the web server 200c according to the present embodiment is compared with the web server 200 according to the first embodiment, the processes of the extraction unit 220c and the filtering processing unit 225c are different. Since the other configuration is the same as that of the first embodiment, the same reference numerals are given and description thereof is omitted, and different portions in the present embodiment will be mainly described.

The season / period acquisition unit 224c has the same function as the season / period acquisition unit 224 (FIG. 10), but differs from the season / period acquisition unit 224 in the following points.
The season / period acquisition unit 224c extracts the question posting date and time when the extracted question was posted from the question information, and refers to the definition information as shown in FIG. Acquired as period information. Further, the season / period acquisition unit 224c extracts the question posting date and time when the extracted question was posted from the question information, and refers to the definition information as shown in FIG. Is acquired as seasonal information. Then, the season / period acquisition unit 224c outputs the acquired period information and season information to the filtering processing unit 225c.

  When there is an answer to the extracted question, the season / period acquisition unit 224c extracts the answer posting date and time when the answer to the question was posted from the question information, and refers to the definition information as shown in FIG. The month of the period to which the answer posting date / time belongs is acquired as period information. Also, the season / period acquisition unit 224c extracts the answer posting date and time from the question information, and acquires the season of the period to which the question posting date and time belongs as season information with reference to the definition information as shown in FIG. To do. Then, the season / period acquisition unit 224c outputs the acquired period information and season information to the filtering processing unit 225c.

The filtering processing unit 225c has the same function as the filtering processing unit 225 (FIG. 10), but differs from the filtering processing unit 225 in the following points.
First, based on the period information input from the season / period acquisition unit 224c, the filtering processing unit 225c extracts the content whose posting date of the extracted content corresponds to the period information. When the extracted content is smaller than the predetermined number, the filtering processing unit 225c is based on the season information input from the season / period acquisition unit 224c, and the content whose posting date / time of the extracted content corresponds to the season information To extract.
In this way, the filtering processing unit 225c extracts the content by changing the filtering period according to the number of contents to be extracted.

  As described above, according to the present embodiment, the information processing apparatus (web server 200c) refers to the preference information indicating the preference for each article of the target user and other users, and the target user and the other users. A similar user extraction unit 221 that calculates a similarity between users, which is a similarity between the two, and extracts a similar user similar to the target user based on the calculated similarity between users, and a similar user extracted by the similar user extraction unit 221 The content extraction unit 222 that extracts the content that the user likes, and the content filtering processing unit (filtering processing unit 225c) that filters the content extracted by the content extraction unit 222 with date and time information and extracts the presentation content to be presented to the target user And comprising.

  As a result, the information processing apparatus (web server 200c) associates content that has recently become a topic or seasonal content with content associated with a period from a plurality of similar users who have similar preferences with a certain user. When the number of extracted contents is extracted every month and the number of extracted contents is small, the number of extracted contents is smaller than a predetermined number. Information that matches the user's current preferences can be presented.

(Sixth embodiment)
Hereinafter, the sixth embodiment of the present invention will be described in detail with reference to the drawings.
Since the information processing system 1e according to the present embodiment is the same as the information processing system 1 according to the first embodiment, description thereof is omitted. Further, the database server 100 and the terminal device 300 according to the present embodiment are the same as the database server 100 and the terminal device 300 according to the first embodiment, and thus description thereof is omitted.

FIG. 24 is a schematic block diagram showing an example of the configuration of the web server 200d according to the sixth embodiment of the present invention.
The web server 200d includes an input / output control unit 210 and an extraction unit 220d. The input / output control unit 210 includes a response control unit 211 and a database reading unit 212. The response control unit 211a includes a character string acquisition unit 214, a display mode acquisition unit 215, a specific content acquisition unit 216, and a category acquisition unit 217. The extraction unit 220d includes a similar user extraction unit 221, a content extraction unit 222, a similar content extraction unit 223, a season / period acquisition unit 224d, and a filtering processing unit 225d. The season / period acquisition unit 224d includes a season / period acquisition unit 2241d from the current time and a browsing history season / period acquisition unit 2242d. The filtering processing unit 225d includes a history filtering processing unit 2251d and an extraction result filtering processing unit 2252d.

  When the web server 200d according to the present embodiment and the web server 200 according to the first embodiment are compared, the processes of the season / period acquisition unit 224d and the filtering processing unit 225d are different. Since the other configuration is the same as that of the first embodiment, the same reference numerals are given and description thereof is omitted, and different portions in the present embodiment will be mainly described.

  The season / period acquisition unit 2241d from the current time acquires information representing the current date and time via a built-in clock or network built in the web server 200d. The season / period acquisition unit 2241d from the current time refers to the definition information as shown in FIG. 14 and specifies and acquires the season to which the current date belongs as season information. Alternatively, the season / period acquisition unit 2241d from the current time refers to the definition information as illustrated in FIG. 14 and specifies and acquires the period to which the current date / time belongs as the period information. For example, if the current date / time is May 1, 2013, the season / period acquisition unit 2241d from the current time is “spring” in FIG. 14 and the period to which the current date / time belongs is In FIG. 14, “March 1 to June 30”. Note that the season / period acquisition unit 2241d from the current time may acquire the period information subdivided according to the time information included in the current date and time with reference to the definition information that defines the time.

  The history filtering processing unit 2251d extracts (filters) content from the database 110 of the database server 100 based on the season information or period information acquired by the season / period acquisition unit 2241d from the current time. Specifically, the history filtering processing unit 2251d is the season or period information in which the question posting date / time and the answer posting date / time are represented by the season information among a plurality of contents (for example, questions) stored in the database 110 of the database server 100. When it is shown that it was posted during the period indicated by, the question posted on the question posting date and the answer posted on the answer posting date are extracted, and the question and answer other than the extracted question and answer are excluded . The history filtering processing unit 2251d outputs the content information representing the extracted question and answer to the similar content extracting unit 223 and the similar user extracting unit 221.

  The similar user extraction unit 221 selects the target user based on the season represented by the season information acquired by the season / period acquisition unit 2241d from the current time or the period represented by the period information and the content information input from the history filtering processing unit 2251d. And similar users with similar preferences for articles. Specifically, the similar user extraction unit 221 extracts other users than the target user who posted or browsed the content information posted in the season represented by the season information or the period represented by the period information. The similar user extraction unit 221 calculates the similarity between users, which is the similarity between the target user and each of the other users in the season represented by the season information or the period represented by the period information. Specifically, the similar user extraction unit 221 calculates the similarity between users using, for example, the above-described equation (1). Based on the calculated similarity between users, the similar user extraction unit 221 extracts other top N users (N is a predetermined number of integers) with high similarity between users as similar users.

  The similar content extraction unit 223 extracts similar content similar to the specific content (for example, a question) acquired by the specific content acquisition unit 216 based on the content information input from the history filtering processing unit 2251d. Specifically, for example, the similar content extraction unit 223 includes a set of users who have already browsed a specific content (hereinafter, also referred to as a first question) included in the content information among the specific content, and the content. The number of users that overlap with a set of users who have already browsed another content (hereinafter also referred to as a second question) is calculated (hereinafter referred to as a user overlap number). And the similar content extraction part 223 calculates the similarity of the 2nd question with respect to a 1st question based on the calculated user duplication number, for example. For example, the similar content extraction unit 223 calculates the similarity of each content with respect to the specific content by calculating the similarity with respect to content other than the second question. Then, the similar content extraction unit 223 extracts, for example, the top M contents (M is a predetermined integer equal to or greater than 1) having a high degree of similarity of the calculated content.

  The content extraction unit 222 extracts the content viewed by the similar user extracted by the similar user extraction unit 221 from the database 110 of the database server 100. Specifically, the content extraction unit 222 extracts content (for example, a question) viewed by the similar user extracted by the similar user extraction unit 221 from the database 110 of the database server 100, and the target user has already browsed from the extracted content. Excluded content That is, the content extraction unit 222 extracts content obtained by excluding content that has already been viewed by the target user from content that has been viewed by similar users.

  The browsing history season / period acquisition unit 2242d acquires the question posting date and the answer posting date and time of the question of the browsing history of the content browsed by the target user. Then, the browsing history season / period acquisition unit 2242d refers to the definition information as illustrated in FIG. 14 and acquires, as history period information, a season or period corresponding to the acquired question posting date / time and answer posting date / time.

  The extraction result filtering processing unit 2252d is based on the season information or period information acquired by the browsing history season / period acquisition unit 2242d from the contents (for example, questions) input by the content extraction unit 222 and the similar content extraction unit 223. To extract (filter) content. Specifically, the filtering result processing unit 2252d of the extraction result indicates that the history period information acquired by the browsing history season / period acquisition unit 2242d among the plurality of input contents is represented by the season or period information represented by the season information. In the case of representing that the posting was made during the period, the question posted at the question posting date and the answer posted at the answer posting date are extracted, and the question and answer other than the extracted question and answer are excluded.

FIG. 25 is a flowchart showing an example of information processing in the web server 200d according to the sixth embodiment of the present invention.
In step ST301, the season / period acquisition unit 2241d from the current time acquires information representing the current date and time, and the season information or information representing the current date and time represents the season to which the current date and time represented by the information representing the current date and time belongs. The period to which the current date belongs is specified and acquired as period information. Then, it progresses to step ST302.
In step ST302, the history filtering processing unit 2251d extracts (filters) content from the database 110 of the database server 100 based on the season information or period information acquired by the season / period acquisition unit 2241d from the current time. Further, the similar user extraction unit 221 calculates the similarity between users between the target user and each of the other users. Then, it progresses to step ST303.

In step ST303, the similar user extracting unit 221 extracts the top N users having the similarities between users as similar users. Then, it progresses to step ST304.
In step ST304, the content extraction unit 222 extracts all questions viewed by similar users. Then, it progresses to step ST305.
In step ST305, the content extraction unit 222 excludes questions already viewed by the target user from all the questions browsed by the similar users extracted in step ST304. Then, it progresses to step ST306.

In step ST306, the browsing history season / period acquisition unit 2242d acquires the history period information of the content viewed by the target user from the browsing history of the content browsed by the target user. Then, it progresses to step ST307.
In step ST307, the filtering processing unit 2252d of the extraction result asks the question posted from the content input by the content extracting unit 222 and the similar content extracting unit 223 during the season indicated by the season information or the period information specified by the user. And answers are extracted, and questions and answers posted outside the period indicated by the season or period information indicated by the season information specified by the user are excluded. Then, it progresses to step ST308.

In step ST308, the extraction result filtering processing unit 2252d calculates a recommendation score (recommendation score) for each of the questions and answers that are left out in step ST307. Then, it progresses to step ST309.
In step ST309, the extraction result filtering processing unit 2252d extracts a predetermined number of questions and answers in descending order of recommendation score (recommendation score), and causes the terminal device 300 to present the extracted questions and answers. Thereafter, the information processing according to FIG. 25 ends.

As described above, according to the present embodiment, the information processing apparatus (web browser 200d) performs the preference information filtering processing unit that filters the preference information indicating the preference for each article of the target user and other users by the date / time information. Referring to the preference information filtered by the (history filtering processing unit 2251d) and the preference information filtering processing unit (history filtering processing unit 2251d), the inter-user similarity that is the similarity between the target user and another user is calculated. Then, based on the calculated similarity between users, a similar user extraction unit 221 that extracts similar users similar to the target user, and a content extraction unit 222 that extracts content preferred by similar users extracted by the similar user extraction unit 221 And comprising.
As a result, the information processing apparatus (web server 200d) allows the content that has recently become a topic or the seasonality from the content viewed by a plurality of similar users who have similar preferences to a certain user filtered by the season information or the period information. You can extract content with

In addition, the information processing apparatus (web server 200d) includes a preference information filtering processing unit (history filtering processing unit 2251d) that filters the preference information indicating the preference of the articles of the target user and other users by date information, Referring to the preference information filtered by the information filtering processor (history filtering processor 2251d), an inter-user similarity that is the similarity between the target user and another user is calculated, and based on the calculated inter-user similarity The similar user extraction unit 221 that extracts similar users similar to the target user, the content extraction unit 222 that extracts the content preferred by the similar user extracted by the similar user extraction unit 221, and the content extraction unit 222 Depending on the date and time information, And Taringu includes content filtering processing unit which extracts the provided content to be presented to the target user (the filtering processing unit 2252d extraction results), the.
As a result, the information processing apparatus (web server 200d) allows the content that has recently become a topic or the seasonality from the content viewed by a plurality of similar users who have similar preferences to a certain user filtered by the season information or the period information. It is possible to extract a certain content, and to present information that matches the user's current preference for filtering the extraction result by season information or period information.

(Seventh embodiment)
Hereinafter, the seventh embodiment of the present invention will be described in detail with reference to the drawings.
Since the information processing system 1f according to the present embodiment is the same as the information processing system 1 according to the first embodiment, description thereof is omitted. Further, the database server 100 and the terminal device 300 according to the present embodiment are the same as the database server 100 and the terminal device 300 according to the first embodiment, and thus description thereof is omitted.
The configuration of the web server 200d according to the seventh embodiment of the present invention will be described with the aid of the configuration of the web server 200d according to the sixth embodiment according to FIG. Note that the configuration of the web server 200d according to the present embodiment is the same as the configuration of the web server 200d according to the sixth embodiment, and thus description thereof is omitted.

FIG. 26 is a flowchart showing an example of information processing in the web server 200d according to the seventh embodiment of the present invention.
In step ST401, the season / period acquisition unit 2241d from the current time acquires information representing the current date and time, and the season information or information representing the current date and time represents the season to which the current date and time represented by the information representing the current date and time belongs. The period to which the current date belongs is specified and acquired as period information. Then, it progresses to step ST402.
In step ST402, the history filtering processing unit 2251d extracts (filters) content from the database 110 of the database server 100 based on the season information or period information acquired by the season / period acquisition unit 2241d from the current time. Further, the similar user extraction unit 221 calculates the similarity between users between the target user and each of the other users. Then, it progresses to step ST403.

In step ST403, the similar user extracting unit 221 extracts the top N users having the similarities between users as similar users. Then, it progresses to step ST404.
In step ST404, the content extraction unit 222 extracts all questions viewed by similar users. Then, it progresses to step ST405.
In step ST405, the content extraction unit 222 excludes questions already viewed by the target user from all the questions browsed by the similar users extracted in step ST304. Then, it progresses to step ST406.

In step ST406, the extraction result filtering processing unit 2252d calculates a recommendation score (recommendation score) for each of the questions and answers that are left out in step ST405. Then, it progresses to step ST407.
In step ST407, the extraction result filtering processing unit 2252d extracts a predetermined number of questions and answers in descending order of recommendation score (recommendation score), and causes the terminal device 300 to present the extracted questions and answers. Then, it progresses to step ST408.
In step ST408, the browsing history season / period acquisition unit 2242d acquires the history period information of the content viewed by the target user from the browsing history of the content browsed by the target user. Then, it progresses to step ST409.

  In step ST409, the filtering processing unit 2252d of the extraction result asks the question posted from the content input by the content extraction unit 222 and the similar content extraction unit 223 during the season indicated by the season information or the period information specified by the user. And answers are extracted, and questions and answers posted outside the period indicated by the season or period information indicated by the season information specified by the user are excluded. Thereafter, the information processing according to FIG. 26 ends.

As described above, according to the present embodiment, the information processing apparatus (web browser 200d) performs the preference information filtering processing unit that filters the preference information indicating the preference for each article of the target user and other users by the date / time information. Referring to the preference information filtered by the (history filtering processing unit 2251d) and the preference information filtering processing unit (history filtering processing unit 2251d), the inter-user similarity that is the similarity between the target user and another user is calculated. Then, based on the calculated similarity between users, a similar user extraction unit 221 that extracts similar users similar to the target user, and a content extraction unit 222 that extracts content preferred by similar users extracted by the similar user extraction unit 221 And comprising.
As a result, the information processing apparatus (web server 200d) allows the content that has recently become a topic or the seasonality from the content viewed by a plurality of similar users who have similar preferences to a certain user filtered by the season information or the period information. You can extract content with

In addition, the information processing apparatus (web server 200d) includes a preference information filtering processing unit (history filtering processing unit 2251d) that filters the preference information indicating the preference of the articles of the target user and other users by date information, Referring to the preference information filtered by the information filtering processor (history filtering processor 2251d), an inter-user similarity that is the similarity between the target user and another user is calculated, and based on the calculated inter-user similarity The similar user extraction unit 221 that extracts similar users similar to the target user, the content extraction unit 222 that extracts the content preferred by the similar user extracted by the similar user extraction unit 221, and the content extraction unit 222 Filter content by date and time information And grayed includes content filtering processing unit which extracts the provided content to be presented to the target user (the filtering processing unit 2252d extraction results), the.
As a result, the information processing apparatus (web server 200d) allows the content that has recently become a topic or the seasonality from the content viewed by a plurality of similar users who have similar preferences to a certain user filtered by the season information or the period information. Some content can be extracted, and information suitable for the current preference of the certain user can be presented.

In each embodiment described above, the specific content acquisition unit 216 may acquire the specific content selected by the target user from all the contents stored in the database server 100.
In each embodiment described above, the similar user extraction unit 221 compares the calculated similarity between users with a predetermined threshold, and selects other users other than the target user whose inter-user similarity is larger than the threshold. You may extract as a similar user.

In each embodiment described above, the content extraction unit 222 includes, for example, content including a character string acquired by the character string acquisition unit 214 from content already browsed by a similar user extracted by the similar user extraction unit 221. You may search and extract the searched content.
In each embodiment described above, the content extraction unit 222 may further extract content classified into the category represented by the category data input from the category acquisition unit 217 from the extracted content. In this case, when the category data input from the category acquisition unit 217 represents a target user's preference information category (for example, a browsing history category), the content extraction unit 222 does not perform a process of further extracting content from the extracted content. To the filtering processing units 225, 225a, 225b, and 225c.

  In each of the above-described embodiments, even if category data is not input from the category acquisition unit 217, the content extraction unit 222 uses the preference information (for example, browsing information) stored in the database 110 to identify the target user. Preference information may be extracted. When the number of dependences (for example, the number of browsing) indicated by the extracted preference information is less than a predetermined frequency, the content extraction unit 222 may output the extracted content to the filtering processing units 225, 225a, 225b, and 225c. In other cases, the content extraction unit 222 may output data indicating the extracted content to the database reading unit 212 as extraction result data.

  Here, when the frequency is less than a predetermined frequency, for example, when the number of reliance (for example, the number of browsing) is less than a predetermined threshold (for example, 10), the last reliance date (for example, the browsing date) ) Is longer than a predetermined elapsed time threshold (for example, one month). In such a case, the information that the target user relies on (accesses) is less reliable as preference information indicating the preference of the target user for the article. That is, when the reliability as the preference information is sufficient, the extracted content is output to the database reading unit 212, and when the reliability is low, the extracted content is output to the filtering processing unit 225.

  In each of the embodiments described above, when the number of relying on the target user (for example, the number of browsing) is 0, the content extracting unit 222 counts the number of relying on other users other than the target user (for example, viewing). The total number of reliabilities (for example, the total number of browsing and the number of hits) may be calculated for each content. In such a case, for example, the target user uses the web server 200 for the first time. The content extraction unit 222 extracts a predetermined number of contents in descending order of the total number of dependences, and outputs extraction result data indicating the extracted content to the database reading unit 212.

  In each embodiment described above, when the category data input from the category acquisition unit 217 represents the preference information category (for example, browsing history category) of the target user, the similar content extraction unit 223 uses the extracted content. Furthermore, you may output to the filtering process part 225, without performing the process to extract.

In each of the above-described embodiments, even if category data is not input from the category acquisition unit 217, the similar content extraction unit 223 determines the target user from preference information (for example, browsing information) stored in the database 110. The dependency information may be extracted. In this case, when the number of dependences (for example, the number of browsing) indicated by the extracted information is less than a predetermined frequency, the similar content extraction unit 223 may output the extracted content to the filtering processing unit 225. Further, when the number of dependences (for example, the number of browsing) indicated by the extracted information is greater than a predetermined frequency, the similar content extraction unit 223 may output data indicating the extracted content to the database reading unit 212 as extraction result data. .
Thereby, when the reliability as the preference information indicating the preference of the object of the target user is sufficient, the extracted content is output to the database reading unit 212, and when the reliability is low, the filtering processing unit 225 Is output.

In each of the above-described embodiments, the filtering processing units 225, 225a, 225b, and 225c may calculate a recommendation score for each of the extracted contents, and extract a predetermined number of contents in descending order of the recommendation score. . In addition, the filtering processing units 225, 225a, 225b, and 225c may calculate a recommendation score for each of the extracted contents, and may extract content that is higher than a predetermined recommendation score in descending order of the recommendation score.
In the sixth embodiment and the seventh embodiment described above, the web servers 200c and 200d may not include the extraction result filtering processing unit 2252d.

  In each of the above-described embodiments, processing related to a question is mainly described as an example of content, but the present invention is not limited to this. If it is a content, it is not restricted to a question, For example, a contribution and an answer may be sufficient. Moreover, although the case where the dependence (access) to the content is mainly browsing has been described, the present invention is not limited thereto, and may be, for example, a post or an answer. Moreover, although the process which mainly concerns browsing information was demonstrated as preference information, if it is the information which shows the preference of articles | goods (for example, content), it will not be restricted to browsing information, For example, contribution information, reply information, and other information Or any combination thereof or all of them.

  In each embodiment described above, the similar user extraction unit 221 calculates the similarity between users based on the preference information indicating the preference of the target user and the content of other users, and uses the calculated similarity between users. Based on this, similar users similar to the target user may be extracted. Here, an example of the preference information indicating the preference for the content is the number of questions viewed. Specifically, for example, the similar user extraction unit 221 determines the number of articles commonly used by the target user and other users × 2 / (total number of contents preferred by the target user + total number of contents preferred by other users. ) May be calculated as the similarity between users. Thereby, the similar user extraction part 221 can extract the user with close preference with respect to a content as a similar user.

  In that case, the content extraction unit 222 may extract the content preferred by the similar user extracted by the similar user extraction unit 221 as the presentation candidate content that is a candidate to be presented to the target user. Thereby, since the content which the similar user with a similar preference with respect to a target user likes can be shown to a target user, the content which a target user likes can be found easily.

  In each of the above-described embodiments, the content extraction unit 222 selects a similar user and a target who likes the similar user preference content for each similar user preference content that is a content preferred by the similar user extracted by the similar user extraction unit 221. A recommendation score indicating a recommendation degree to the target user of the similar user preference content is calculated with reference to the similarity between users between the users, and presentation candidate content is determined based on each of the calculated recommendation scores May be. At that time, the content extraction unit 222 may extract, as the presentation candidate content, content excluding content preferred by the target user from content preferred by similar users.

Moreover, in each embodiment mentioned above, although the content extraction part 222 demonstrated taking as an example the case where the target user browsed the question browsed even once from a presentation candidate question, the target user decided a predetermined number of times (for example, (3 times) The browsed questions may be excluded from the presentation candidate questions.
In each embodiment described above, the content extraction unit 222 may exclude questions asked by similar users in the past for a predetermined period (for example, one year) from the present from the presentation candidate questions.

  In each of the above-described embodiments, the case where the content extraction unit 222 mainly extracts questions browsed by similar users in the past has been described as an example, but the present embodiment is not limited to this. The content extraction unit 222 may extract, for example, any of content previously browsed by similar users, posted content, and answered content, any combination thereof, or all of them.

  Moreover, in each embodiment mentioned above, the similar content extraction part 223 may extract the similar content similar to the specific content which the specific content acquisition part 216 acquired. As a result, the information processing systems 1, 1 a, 1 b, 1 c, 1 d, 1 e, and 1 f can present content similar to the specific content selected by the target user to the target user. Find content quickly and easily. At that time, for example, the similar content extraction unit 223 refers to the number of users who prefer the specific content and the number of users who prefer each of the presentation candidate contents, and the specific content and the presentation candidate content. The content similarity with each may be calculated, and similar content may be extracted based on the calculated content similarity.

  Thereby, the similar content extraction part 223 can make the content similarity between the content high, so that there are many users who like and overlap between specific content and presentation candidate content. The similar content extraction unit 223 considers that the higher the content similarity is, the more similar the content is. For example, content with the highest content similarity M (M is an integer greater than a predetermined 0) is set as similar content. Can be extracted. As a result, the similar content extraction unit 223 can extract similar content with high accuracy.

  In each of the above-described embodiments, the similar content extraction unit 223 may extract, as similar content, content whose content similarity is higher than a predetermined content similarity criterion, for example.

In addition, in each embodiment mentioned above, you may combine with each other embodiment, for example, you may combine like 1st Embodiment and 3rd Embodiment.
In each of the above-described embodiments, an example in which the date and time information is seasonal information, keyword information, and period information has been described. However, time information representing time, period information representing day, period information representing year, and the like. Date and time information regarding various times and periods may be included. Further, although an example in which the keyword information is a keyword related to the season has been shown, it may be a keyword related to time, a period, etc., such as a keyword related to time, a keyword indicating day, and a keyword indicating month.

  In each embodiment described above, the information processing systems 1, 1a, 1b, 1c, 1d, 1e, and 1f may include a plurality of web servers 200, 200a, 200b, 200c, and 200d. These web servers 200, 200a, 200b, 200c, and 200d may perform their respective processes in a distributed manner.

  Note that some or all of the web servers 200, 200a, 200b, 200c, and 200d in each of the above-described embodiments may be realized by a computer. In that case, the program for realizing the control function may be recorded on a computer-readable recording medium, and the program recorded on the recording medium may be read by a computer system and executed. Here, the “computer system” is a computer system built in the web servers 200, 200a, 200b, 200c, and 200d, and includes hardware such as an OS (Operating System) and peripheral devices. Further, the “computer system” includes a homepage providing environment (or display environment) if a WWW (World Wide Web) system is used.

  The “computer-readable recording medium” refers to a portable disk such as a flexible disk, a magneto-optical disk, a ROM (Read-only Memory), a writable nonvolatile memory such as a flash memory, a CD (Compact Disc) -ROM, or the like. A storage device such as a medium or a hard disk built in a computer system. Furthermore, the “computer-readable recording medium” is a medium that dynamically holds a program for a short time, such as a communication line when transmitting a program via a network such as the Internet or a communication line such as a telephone line, A volatile memory (for example, DRAM (Dynamic Random Access Memory)) inside a computer system that serves as a server or a client in that case may be included.

  The program may be a program for realizing a part of the functions described above, and may be a program capable of realizing the functions described above in combination with a program already recorded in a computer system. The program may be transmitted from a computer system storing the program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in the transmission medium. Here, the “transmission medium” for transmitting the program refers to a medium having a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line. The program may be for realizing a part of the functions described above. Furthermore, what can implement | achieve the function mentioned above in combination with the program already recorded on the computer system, what is called a difference file (difference program) may be sufficient.

  Moreover, you may implement | achieve part or all of the web servers 200, 200a, 200b, 200c, and 200d in each embodiment mentioned above as integrated circuits, such as LSI (Large Scale Integration). Each functional block of the web servers 200, 200a, 200b, 200c, and 200d may be individually made into a processor, or a part or all of them may be integrated into a processor. Further, the method of circuit integration is not limited to LSI, and may be realized by a dedicated circuit or a general-purpose processor. Further, in the case where an integrated circuit technology that replaces LSI appears due to progress in semiconductor technology, an integrated circuit based on the technology may be used.

  As mentioned above, although each embodiment of this invention was explained in full detail with reference to drawings, a specific structure is not restricted to embodiment. Each configuration in each embodiment, a combination thereof, and the like are examples, and addition, omission, replacement, and other changes of the configuration can be made without departing from the spirit of the present invention. Further, the present invention is not limited by each embodiment, and is limited only by the scope of the claims.

  (Supplementary Note 1) With reference to preference information filtered by the preference information filtering processing unit and preference information filtering processing unit that filters preference information indicating preference for each article of the target user and other users by date and time information. A similar user extraction unit that calculates a similarity between users, which is a similarity between the target user and the other user, and extracts a similar user similar to the target user based on the calculated similarity between the users; An information processing apparatus comprising: a content extraction unit that extracts content that the similar user extracted by the similar user extraction unit likes.

  (Additional remark 2) With reference to the preference information which shows the preference with respect to each article | item of a target user and another user, the similarity degree between users which is a similarity degree of the said target user and the said other user was calculated, and it calculated A similar user extraction unit that extracts similar users similar to the target user based on the similarity between the users, a content extraction unit that extracts content that the similar user extracted by the similar user extraction unit likes, An information processing apparatus comprising: a content filtering processing unit that filters the content extracted by a content extraction unit by date information and extracts a presentation content to be presented to the target user.

  (Additional remark 3) With reference to the preference information filtered by the preference information filtering process part and the preference information filtering process part which filters the preference information which shows the preference with respect to each goods of an object user and another user by date information, and the preference information filtering process part A similar user extraction unit that calculates a similarity between users, which is a similarity between the target user and the other user, and extracts a similar user similar to the target user based on the calculated similarity between the users; A content extraction unit that extracts the content that the similar user likes that is extracted by the similar user extraction unit, and the presented content that is presented to the target user by filtering the content extracted by the content extraction unit by date information A content filtering processing unit for extracting The information processing apparatus according to claim and.

  (Supplementary Note 4) The preference information indicating the preference for the article is history information including at least one of a browsing history of content including at least one of a question and an answer or a posting history of the content, and the date and time information includes content The information processing apparatus according to any one of (Appendix 1) to (Appendix 3), wherein the information is at least one of date and time information browsed and date and time information posted.

  (Supplementary Note 5) The preference information indicating the preference for the article is purchase history of the article, and the date / time information is date / time information when the content is purchased (Appendix 1) to (Appendix 3) The information processing apparatus according to any one of claims.

  (Additional remark 6) It further has the keyword extraction part which extracts a keyword from the said content extracted by the said content extraction part, The said date information is the date information previously matched with the said keyword extracted by the said keyword extraction part. The information processing apparatus according to any one of (Appendix 1) to (Appendix 3).

  (Additional remark 7) With reference to the preference information filtered by the preference information filtering process part and the preference information filtering process part which filters the preference information which shows the preference with respect to each goods of an object user and another user by date information, and the said preference information filtering process part A similar user extraction unit that calculates a similarity between users, which is a similarity between the target user and the other user, and extracts a similar user similar to the target user based on the calculated similarity between the users; An information processing system comprising: a content extraction unit that extracts content that the similar user extracted by the similar user extraction unit likes.

  (Additional remark 8) With reference to the preference information which shows the preference with respect to each article | item of a target user and another user, the similarity degree between users which is the similarity degree of the said target user and the said other user was calculated, and it calculated A similar user extraction unit that extracts similar users similar to the target user based on the similarity between the users, a content extraction unit that extracts content that the similar user extracted by the similar user extraction unit likes, An information processing system comprising: a content filtering processing unit that filters the content extracted by a content extraction unit with date and time information and extracts presentation content to be presented to the target user.

  (Additional remark 9) With reference to the preference information filtered by the preference information filtering process part and the preference information filtering process part which filters the preference information which shows the preference with respect to each article | item of an object user and another user with date information A similar user extraction unit that calculates a similarity between users, which is a similarity between the target user and the other user, and extracts a similar user similar to the target user based on the calculated similarity between the users; A content extraction unit that extracts the content that the similar user likes that is extracted by the similar user extraction unit, and the presented content that is presented to the target user by filtering the content extracted by the content extraction unit by date information A content filtering processing unit for extracting An information processing system which is characterized the door.

  (Additional remark 10) With reference to the preference information filtered by the preference information filtering process in the preference information filtering process in which the preference information indicating the preference for each article of the target user and other users is filtered by date and time information Calculating a similarity between users, which is a similarity between the target user and the other users, and extracting a similar user similar to the target user based on the calculated similarity between the users; A content extraction process for extracting content that the similar user likes and extracted by the similar user extraction process.

  (Additional remark 11) With reference to the preference information which shows the preference with respect to each article | item of a target user and another user, the similarity degree between users which is the similarity degree of the said target user and the said other user was calculated, and it calculated A similar user extraction process for extracting similar users similar to the target user based on the similarity between the users, a content extraction process for extracting content preferred by the similar users extracted by the similar user extraction process, An information processing method comprising: a content filtering processing step of filtering the content extracted in the content extraction step by date information and extracting a presentation content to be presented to the target user.

  (Additional remark 12) With reference to the preference information filtered by the preference information filtering process process and the preference information filtering process process which filters the preference information which shows the preference with respect to each article | item of an object user and another user by date information Calculating a similarity between users, which is a similarity between the target user and the other users, and extracting a similar user similar to the target user based on the calculated similarity between the users; A content extraction process for extracting content preferred by the similar user extracted by the similar user extraction process, and a content to be presented to the target user by filtering the content extracted by the content extraction process by date information Filtering process to extract The information processing method characterized by comprising the extent, the.

  (Additional remark 13) It is filtered by the preference information filtering process step which filters the preference information which shows the preference with respect to each goods of a target user and another user by the date information on the computer of information processing apparatus, and the said preference information filtering process step Referring to the preference information, the similarity between the users, which is the similarity between the target user and the other user, is calculated, and similar users similar to the target user are extracted based on the calculated similarity between the users An information processing program for executing a similar user extraction step and a content extraction step for extracting content that the similar user extracted by the similar user extraction step likes.

  (Additional remark 14) Refer to the preference information which shows the preference with respect to each article | item of a target user and another user to the computer of information processing apparatus, and the similarity between users which is the similarity degree of the said target user and the said other user A similar user extraction step of extracting a similar user similar to the target user based on the calculated similarity between the users, and extracting the content preferred by the similar user extracted by the similar user extraction step An information processing program for executing a content extracting step for performing, and a content filtering processing step for filtering the content extracted by the content extracting step based on date and time information to extract a presentation content to be presented to the target user.

  (Additional remark 15) It is filtered by the preference information filtering process step which filters the preference information which shows the preference with respect to each goods of an object user and another user by the date information on the computer of information processing apparatus, and the said preference information filtering process step Referring to the preference information, the similarity between the users, which is the similarity between the target user and the other user, is calculated, and similar users similar to the target user are extracted based on the calculated similarity between the users The similar user extraction step, the content extraction step for extracting the content that the similar user extracted by the similar user extraction step, the content extracted by the content extraction step is filtered by date information, Present to the target user The information processing program for executing a content filtering processing step of extracting the present content and.

1, 1a, 1b, 1c, 1d, 1e, 1f ... information processing system, 100 ... database server, 110 ... database, 120 ... communication unit, 130 ... CPU, 200, 200a, 200b, 200c, 200d ... Web server (information processing apparatus), 201 ... CPU, 202 ... storage medium, 203 ... drive unit, 204 ... input unit, 205 ... output unit, 206 ... ROM, 207 ... RAM, 208 ... auxiliary storage unit, 209 ... communication unit, 210 ... input / output control unit, 211, 211a, 211b ... response control unit, 212. .. database reading unit, 214... Character string acquisition unit, 215... Display mode acquisition unit, 216... Specific content acquisition unit, 217. , 220b, 220c, 220d ... extraction unit, 221 ... similar user extraction unit, 222 ... content extraction unit, 223 ... similar content extraction unit, 224, 224a, 224b, 224c, 224d ... Season / period acquisition unit, 225, 225a, 225b, 225c, 225d ... Filtering processing unit, 2241d ... Season / period acquisition unit from current time, 2242d ... Season / period acquisition unit of browsing history, 2251d ... History filtering processing unit (preference information filtering processing unit), 2252d ... Extraction result filtering processing unit (content filtering processing unit), 300 ... Terminal device, 310 ... Storage unit, 320 ... Communication unit, 330 ... CPU, 340 ... display unit

The present invention relates to an information processing device, an information processing system, an information processing method , an information processing program, and a terminal device .

According to the information processing device, the information processing system, the information processing method , the information processing program, and the terminal device according to one aspect of the present invention, it is possible to present information that meets the user's preference.

Claims (9)

A preference information filtering processing unit that filters preference information indicating preference for each article of the target user and other users by date and time information;
With reference to the preference information filtered by the preference information filtering processing unit, the similarity between the users that is the similarity between the target user and the other user is calculated, and based on the calculated similarity between the users A similar user extraction unit that extracts similar users similar to the target user;
A content extraction unit that extracts the content that the similar user likes, extracted by the similar user extraction unit;
An information processing apparatus comprising: Referring to the preference information indicating the preference for each article of the target user and other users, the similarity between the users, which is the similarity between the target user and the other users, is calculated, and the calculated similarity between the users A similar user extraction unit that extracts similar users similar to the target user based on the degree;
A content extraction unit that extracts the content that the similar user likes, extracted by the similar user extraction unit;
A content filtering processing unit that filters the content extracted by the content extraction unit according to date and time information and extracts presentation content to be presented to the target user;
An information processing apparatus comprising: A preference information filtering processing unit that filters preference information indicating preference for each article of the target user and other users by date and time information;
With reference to the preference information filtered by the preference information filtering processing unit, the similarity between the users that is the similarity between the target user and the other user is calculated, and based on the calculated similarity between the users A similar user extraction unit that extracts similar users similar to the target user;
A content extraction unit that extracts the content that the similar user likes, extracted by the similar user extraction unit;
A content filtering processing unit that filters the content extracted by the content extraction unit according to date and time information and extracts presentation content to be presented to the target user;
An information processing apparatus comprising: The preference information indicating the preference for the article is history information including at least one of a browsing history of content including at least one of a question or an answer or a posting history of the content,
The information processing apparatus according to any one of claims 1 to 3, wherein the date / time information is at least one of date / time information when content is browsed and date / time information posted. The preference information indicating the preference for the product is a purchase history of the product,
The information processing apparatus according to any one of claims 1 to 3, wherein the date information is date information when content is purchased. A keyword extraction unit for extracting a keyword from the content extracted by the content extraction unit;
The information processing apparatus according to any one of claims 1 to 3, wherein the date / time information is date / time information associated in advance with the keyword extracted by the keyword extraction unit. A preference information filtering processing unit that filters preference information indicating preference for each article of the target user and other users by date and time information;
With reference to the preference information filtered by the preference information filtering processing unit, the similarity between the users that is the similarity between the target user and the other user is calculated, and based on the calculated similarity between the users A similar user extraction unit that extracts similar users similar to the target user;
A content extraction unit that extracts the content that the similar user likes, extracted by the similar user extraction unit;
An information processing system comprising: A preference information filtering process for filtering preference information indicating preference for each article of the target user and other users by date and time information;
With reference to the preference information filtered by the preference information filtering process, the similarity between the users, which is the similarity between the target user and the other user, is calculated, and based on the calculated similarity between the users, A similar user extraction process for extracting similar users similar to the target user;
A content extraction process for extracting content preferred by the similar user extracted by the similar user extraction process;
An information processing method characterized by comprising: In the computer of the information processing device,
A preference information filtering processing step of filtering preference information indicating preference for each article of the target user and other users by date and time information;
With reference to the preference information filtered by the preference information filtering processing step, the similarity between the users that is the similarity between the target user and the other user is calculated, and based on the calculated similarity between the users A similar user extraction step of extracting similar users similar to the target user;
A content extraction step of extracting content that the similar user extracted by the similar user extraction step prefers;
Information processing program to execute.

Images