JP2024076107A - Search support device and search support method - Google Patents

Abstract

[Problem] To enable efficient and accurate information searches that appropriately take into account the search habits and individual attributes of searchers. [Solution] A search support device 100 includes a storage device 101 that holds attribute information and search history of each searcher who has performed an information search, and dictionary information related to fields that may be the subject of a search, and a calculation device 104 that compares a search keyword contained in a new search request with the dictionary information, identifies vocabulary that can be considered to be the same as or synonymous with the search keyword, and, based on the attribute information and search history, identifies search histories by other searchers that include the searcher attributes contained in the search request as attributes and that use the identified vocabulary as a search keyword, and responds with information on the search results indicated by the search history. [Selected Figure] Figure 1

Description

The present invention relates to a search support device and a search support method.

People who search for information on the Internet (hereafter referred to as "searchers") enter keywords that come to mind into a search engine and view the search results. However, in many cases, whether the search results provide the desired information depends on the appropriateness of the keywords.

For example, a conventional technique for making information searches more efficient involves an information presentation device suitable for application to an interactive information search system that follows and presents information with a hypertext structure in response to user instructions, and in particular has been proposed that analyzes trends in the presentation order from the presentation history of information, making it possible to efficiently guide the user to the desired information (see Patent Document 1).

This technology relates to an information presentation device that holds a plurality of pieces of information and presents the information in accordance with a user's instructions, and that is characterized by comprising a presentation history recording means for recording the presentation history of the information, a candidate information extraction means for analyzing the presentation history stored in the presentation history recording means and extracting, for each piece of information, information that is likely to be presented after the previous piece as candidate information, a candidate information storage means for storing the candidate information extracted by the candidate information extraction means, and a request input means for presenting a list of the candidates stored by the candidate information storage means and allowing the user to select which candidate to present from the presented list, and allowing the user to use the presentation order tendency obtained from the presentation history of the information.

In addition, because queries are generally not sent to search engines until the user signals that the query is complete, techniques have been proposed to speed up the process by which time passes while a user completes the full search query (see Patent Document 2).

The technology relates to a method for suggesting query completion for a language having ideographic characters and phonetic characters, the method including the steps of receiving from a search requester a partial query that includes zero or more ideographic characters followed by at least one phonetic character, predicting from the partial query one of a set of predicted queries ordered according to a ranking criterion, the set including the ideographic characters, and communicating the set of ordered predicted queries to the search requester.

Japanese Patent Application Laid-Open No. 9-244945 JP 2008-520037 A

Although various technologies have been proposed, no function is provided for search control based on the suitability of keywords selected by searchers. As a result, searchers are forced to repeatedly select inappropriate keywords, i.e., keywords that do not efficiently lead to desired search results. Of course, there is no function for appropriately reflecting the attributes of individual or overall searchers in such control, and it is difficult to expect improvements in search efficiency and accuracy for searchers.

The object of the present invention is to provide technology that enables efficient and accurate information retrieval that appropriately takes into account the search habits and individual attributes of the searcher.

The search support device of the present invention that solves the above problem includes a storage device that stores attribute information and search history of each searcher who performed an information search, and dictionary information on fields that may be the subject of a search, and when a new search request is received from a terminal, a process that compares the search keyword contained in the search request with the dictionary information and identifies vocabulary that can be considered to be the same as or synonymous with the search keyword, a process that identifies, based on the attribute information and search history, search histories by other searchers that include the searcher attribute contained in the search request as an attribute and that use the identified vocabulary as a search keyword, and a calculation device that executes a process of responding to the terminal with information on the search results indicated by the identified search history.

The search support method of the present invention is characterized in that an information processing device includes a storage device that stores attribute information and search history of each searcher who performed an information search, and dictionary information related to fields that may be the subject of a search, and when a new search request is received from a terminal, the information processing device executes a process of comparing the search keyword contained in the search request with the dictionary information and identifying vocabulary that can be considered to be the same as or synonymous with the search keyword, a process of identifying, based on the attribute information and the search history, search histories by other searchers that include the searcher attribute contained in the search request as an attribute and that use the identified vocabulary as a search keyword, and a process of responding to the terminal with information on the search results indicated by the identified search history.

The present invention enables efficient and accurate information retrieval that appropriately takes into account the search habits and individual attributes of the searcher.

1 is a diagram illustrating a network configuration including a search support device according to an embodiment of the present invention. FIG. 2 is a diagram illustrating an example of a hardware configuration of a search support device according to the present embodiment. FIG. 2 is a diagram illustrating an example of the configuration of an attribute information DB in the present embodiment. FIG. 2 is a diagram illustrating an example of a configuration of a search history DB in the present embodiment. FIG. 2 is a diagram illustrating an example of the configuration of a dictionary DB according to the present embodiment. FIG. 2 is a diagram illustrating an example of the configuration of a habit information DB according to the present embodiment. FIG. 2 is a diagram showing an example of a flow of a search support method according to the present embodiment. FIG. 4 is a diagram showing an example of a screen in the present embodiment. FIG. 4 is a diagram showing an example of a screen in the present embodiment.

<Network Configuration>
An embodiment of the present invention will be described in detail below with reference to the drawings. Fig. 1 is a diagram showing a network configuration including a search support device 100 according to the present embodiment. The search support device 100 shown in Fig. 1 is a computer that enables efficient and accurate information search that appropriately takes into account the search habits and individual attributes of a searcher.

As shown in FIG. 1, the search support device 100 of this embodiment is communicatively connected to a user terminal 200 operated by each searcher via an appropriate network 1 such as the Internet. Therefore, these may be collectively referred to as a search support system 10.

The search support device 100 of this embodiment is a device that provides search services in the network 1, and is also a support device that optimizes searches there.

On the other hand, the user terminal 200 is a terminal used by a searcher to search for information. Specifically, the user terminal 200 may be a smartphone, a tablet terminal, a personal computer, or the like that can connect to the network 1 and display a search UI (User Interface) such as a browser.
<Hardware Configuration>
The hardware configuration of the search support device 100 of this embodiment is as shown in FIG.

That is, the search support device 100 includes a storage device 101, a memory 103, a computing device 104, and a communication device 105.

Of these, the storage device 101 is composed of an appropriate non-volatile storage element such as an SSD (Solid State Drive) or a hard disk drive.

The memory 103 is also composed of a volatile memory element such as a RAM (Random Access Memory).

The calculation device 104 is a CPU (Central Processing Unit) that executes the program 102 stored in the storage device 101 by reading it into the memory 103, thereby controlling the device itself and performing various judgments, calculations, and control processes.

The communication device 105 is assumed to be a network interface card or the like that is connected to the network 1 and handles communication processing with the user terminal 200.

In addition, if the search support device 100 is a search server itself that not only accepts search requests from the user terminal 200 via the network 1 but also can operate as a standalone machine, it is preferable for it to further include an input device as a console that directly accepts key input or voice input from the user, and an output device such as a display that displays processed data.

In addition to the program 102 for implementing the functions required for the search support device 100 of this embodiment, at least a search engine 1021, an attribute information DB 125, a search history DB 126, a dictionary DB 127, and a habit information DB 128 are stored in the storage device 101. However, details of these various databases will be described later.
<Data structure example>
Next, a description will be given of various types of information used by the search support device 100 of this embodiment.

The attribute information DB125 in this embodiment is a database that stores attribute information of each searcher who has previously performed a search on a specific search site, etc.

This attribute information DB125 is a collection of records that use, for example, identification information (e.g., ID, name, etc.) that uniquely identifies a searcher as a key, and link values such as the searcher's age, sex, location, purchasing tendency for specific products and services (which may include amounts of financial product purchases and investments, risk tolerance tendencies, loan amounts and tendencies, etc.), and store usage tendencies.

It is assumed that records including such values are generated from information included in a search request obtained from the user terminal 200 when the search support device 100 operates as a server providing a search service.

Of course, the information may include not only information obtained from the user terminal 200, but also information obtained from the systems of affiliated financial institutions, etc. In this case, it is assumed that the financial institutions, etc. have obtained prior consent from the searcher to provide information about the searcher to the search support device 100.

Next, FIG. 4 shows an example of the search history DB 126 in this embodiment. The search history DB 126 in this embodiment is a database that stores the search contents of each searcher who has performed a search on a specific search site or the like in the past.

This search history DB126 is a collection of records that link the execution date and time and content of each search request issued in the past by the searcher with the search results, using, for example, identification information (e.g., ID or name) that uniquely identifies the searcher as a key. The content of the search request is the search keyword entered by the searcher, and the search results include the URL of the website returned by the search engine 1021 using the search keyword.

In this embodiment, an example has been given in which the attribute information DB 125 and the search history DB 126 are operated as separate databases, but these databases may be operated as a single database by merging records using the searcher as a key.

Next, FIG. 5 shows an example of the dictionary DB 127 in this embodiment. The dictionary DB 127 in this embodiment is a database that stores dictionary information about fields that may be searched by the above-mentioned searcher, in this case, financial products.

This dictionary DB127 is a collection of records that link values such as the official name of the financial product, variations of the official name, and conceptual names higher or lower than the official name, using, for example, identification information (e.g., ID or product name) that uniquely identifies the financial product as a key.

As an example of the above-mentioned "variation in official name," if the "official name" is "XYZ refreshing housing loan," possible variations include "XYZ refreshing," "XYZ housing loan," "refreshing housing loan," "XYZ mortgage," "refreshing housing loan," "refreshing loan," etc. Note that "variation" may also include the concept of descriptions that are considered to be typos or misidentifications.

Among the above-mentioned "higher or lower concept names of the official name," examples of higher concept names include "refreshing loan" and "XYZ loan." Examples of lower concept names include "XYZ refreshing fixed-rate housing loan" and "XYZ refreshing variable-rate housing loan."

Next, FIG. 6 shows an example of the habit information DB 128 in this embodiment. The habit information DB 128 in this embodiment is a database that stores the transitions of a series of search keywords selected by each searcher until the final search result (e.g., a website that the searcher viewed for a certain period of time or more after finishing inputting a series of search keywords).

This habit information DB128 is a collection of records that link data such as the chain of search keywords resulting from the search and the search results obtained at the final stage of the chain, using, for example, identification information (e.g., ID or name) that uniquely identifies the searcher as a key.

A specific example of the above-mentioned "search keyword chain" would be a chain of search keywords entered by the same searcher within a certain period of time, such as "mortgage screening, interest rate" -> "mortgage, borrowing limit" -> "mortgage, borrowing amount, maximum" -> "mortgage, income, borrowing amount" -> "mortgage, borrowing amount, interest rate."

A specific example of the above-mentioned "search result" may be "XX Bank - Simulation of available loan amount."
<Flow example>
The actual procedure of the search support method in this embodiment will be described below with reference to the drawings. The various operations corresponding to the search support method described below are realized by a program that is read into a memory or the like and executed by the search support device 100. This program is composed of codes for performing the various operations described below.

Figure 7 is a diagram showing an example of the flow of the search support method in this embodiment. In this case, it is assumed that the search support device 100 receives a new search request from the user terminal 200 (s10).

The search support device 100 delivers an input screen (see screen 1000 in FIG. 8) for accepting such search requests to the user terminal 200, and accepts the search requests through the input screen.

As shown in the example of FIG. 8, it is preferable that the input screen includes an attribute input field 802 in addition to a search keyword input field 801. The attribute input field 802 shown in FIG. 8 includes input fields for items such as the searcher's age, sex, investment history, residential area, current loan amount, investment risk tolerance, savings amount, and frequently used banks.

Therefore, when accepting a search request, the search support device 100 can also acquire attribute information of the searcher. The search support device 100 generates a record including the acquired attribute information, associates this with the identification information of the searcher, and stores it in the attribute information DB 125.

Next, the search support device 100 checks the search keywords contained in the search request received in s10 against the dictionary DB 127 to identify vocabulary that can be considered to be the same as or synonymous with the search keywords (s11).

Let us assume that the search keywords obtained in s10 are a set of "XY fixed mortgage type, loan amount, loan interest rate." Similarly, let us assume that the attribute information obtained in s10 for the searcher is age "35 years old," gender "male," investment experience "5 years," residential area "Minato Ward, Tokyo," current loan amount "scholarship amount 1.5 million yen," investment risk tolerance "5 million yen," savings amount "20 million yen," and frequently used bank "XYZ Bank."

In this case, the search support device 100 compares each of the search keywords included in the above-mentioned search request, "XY fixed type housing loan, loan amount, loan interest rate," with the dictionary DB 127, and identifies matches with each of "XY fixed type housing loan," "loan amount," and "loan interest rate" in at least one of the formal name, fluctuation, and higher-level or lower-level concept name.

For example, for "XY fixed-rate housing loan," "XYZ Bank housing loan, fixed-rate type" in dictionary DB127 is identified as the official name of the financial product. Furthermore, for "loan amount," "available loan amount" is identified as a variation. Furthermore, for "loan interest rate," "interest rate" is identified as the higher-level concept name.

As described above, when identifying vocabulary that takes "fluctuations" into account or identifying "higher or lower" concept names, it is preferable to adopt an operation that selects and identifies other keywords that are set with those keywords, i.e., keywords that are likely to be linked to the "home loan" corresponding to the product type (predetermined in advance in dictionary DB127), such as "XY fixed type home loan" in the above case.

If the product type is "home loan," a specific example would be to identify "available loan amount" within the "fluctuation" range as a term related to "borrowing" that is easily linked to this, and to identify "interest rate" as the higher-level conceptual name of "loan interest rate."

Next, the search support device 100 compares the attribute information included in the search request received in s10, i.e., the information on the searcher attributes, or the information on the searcher attributes identified in the attribute information DB 125 based on the identification information of the searcher who issued the search request, with the attribute information of other searchers in the attribute information DB 125, and identifies search histories by other searchers that include the searcher attributes as attributes and that use the vocabulary identified in s11 as a search keyword (s12).

In the above example, the search support device 100 identifies the search history of other searchers in the search history DB 126, which includes at least the values of items that are likely to influence the purchase of financial products, etc. (obviously, this is merely an example and is not limiting), such as the searcher's age, residential area, current loan amount, and frequently used bank, from among the searcher attributes obtained about the searcher, such as age "35 years old," gender "male," investment history "5 years," residential area "Minato Ward, Tokyo," current loan amount "1.5 million yen scholarship," investment risk tolerance amount "5 million yen," savings amount "20 million yen," and frequently used bank.

Then, from among the records identified here, those using the search keyword "Home loan" that is included in at least the product name "XYZ Bank Home Loan" among the vocabulary identified in s11, i.e., "XYZ Bank Home Loan Fixed Interest Rate," "Loanable Amount," and "Interest Rate," are identified.

The reason why "Home Loan" was specified here, rather than the official name of the financial product, "XYZ Bank Home Loan", is because we intend to form a certain size population without narrowing the search range of the search history too much. Of course, there is no problem if we can identify the search history by the official name and identify a certain number of search histories.

Therefore, when identifying the above-mentioned search history, the search support device 100 preferably identifies search histories by other searchers that include the searcher attributes included in the search request as attributes, in which the vocabulary identified in s11 is used as a search keyword, and that show an occurrence frequency above a certain standard. An example of such an occurrence frequency can be the most frequent occurrence within a certain period of time, such as the past year or month.

In addition, when identifying the search history based on the searcher attributes in step S12 described above, as another example of the searcher attributes, an operation may be adopted in which the search history is identified using the habit information stored in the habit information DB128 for the searcher as the searcher attributes.

In this case, the search support device 100 searches the habit information DB 128 for records that include the search keyword included in the search request at any stage of the series of search keywords (for example, those at the time of the first input). At this time, if search requests from the searcher have been received multiple times in succession, the search support device 100 checks the chain of search keywords indicated by the successively received search requests against each record in the habit information DB 128, and identifies records in which the chain matches or matches to a certain extent or more as the search history to be identified in s12.

Next, the search support device 100 responds to the user terminal 200 with information about the search results indicated by the search history identified in s12 (s13), and ends the process.

The search support device 100 may execute a search using an existing search engine 1021 based only on the search keywords received in s10. In this case, the search support device 100 may execute control in response to s13 to list and display on the same screen (see screen 1010 in FIG. 9 ) the information on the search results indicated by the search history identified up to s12 (the "search results after correction" in the figure) and the search results by the above-mentioned search engine 1021 (the "search results before correction" in the figure).

If this type of operation were to be implemented, it would be possible to present to the searcher search results based on the (not very suitable) search keywords of the current searcher, i.e., search results from existing general search engines, in a form that compares the search results when the search support technology of the present invention is applied. This would enable the searcher to recognize his or her own search habits, etc., and obtain guidelines for selecting appropriate search keywords in the future.

The above describes in detail the best mode for carrying out the present invention, but the present invention is not limited to this, and various modifications are possible without departing from the spirit of the present invention.

This embodiment enables efficient and accurate information searches that appropriately take into account the search habits and individual attributes of the searcher.

The description in this specification makes clear at least the following. That is, in the search support device of this embodiment, when identifying the search history, the calculation device may identify, based on the attribute information and the search history, search histories by other searchers that include the searcher attribute included in the search request as an attribute, which use the identified vocabulary as a search keyword and show an occurrence frequency above a certain standard.

This makes it possible to avoid the influence of the current searcher's own habits, i.e., the habits of selecting search keywords that are not particularly common, and to utilize the search history of, for example, the majority of other searchers using search keywords, and present that information to the current searcher. Ultimately, it makes possible more efficient and accurate information retrieval that appropriately takes into account the search habits and individual attributes of the searcher.

The storage device may store, as the dictionary information, information on the name of a product or service, a variation of the name, and a higher or lower conceptual name of the name, and when identifying the vocabulary, the computing device may compare a search keyword included in the search request with the dictionary information and identify, as the vocabulary, the name of a product or service that matches the search keyword with at least one of the name, the variation, and the conceptual name.

This makes it possible to avoid the influence of the current searcher's own habits and knowledge, i.e., the tendency to use abbreviations, nicknames, or incorrect names as search keywords rather than the official names of the products or services being searched for, and to utilize the search history of other searchers and present that information to the current searcher. Ultimately, it becomes possible to perform more efficient and accurate information searches that appropriately take into account the search habits and individual attributes of the searcher.

Furthermore, in the search support method of this embodiment, the storage device may hold, as the attribute information, information regarding at least one of the age, gender, location, purchasing tendency of specified products or services, and store usage tendency of each of the searchers, and the calculation device, when identifying the search history, may identify search histories by other searchers that contain, as attributes, searcher attributes that are at least one of the age, gender, location, purchasing tendency of specified products or services, and store usage tendency of the searcher included in the search request, and that use the identified vocabulary as a search keyword.

This makes it possible to use the search history of other searchers with the same age, sex, place of residence, etc., in addition to handling search keywords, and present that information to the current searcher. Ultimately, this enables more efficient and accurate information searches that appropriately take into account the search habits and individual attributes of the searcher.

The search support device in this embodiment may use the computing device to execute a search using an existing search engine based only on the search keyword, and when responding, may respond with information about the search results indicated by the search history together with the results of the search.

This allows the searcher to compare search results based on the (not very suitable) search keywords of the current searcher, i.e., search results from an existing general search engine, with search results when the search support technology of the present invention is applied, allowing the searcher to recognize his or her own search habits, etc., and obtain guidelines for selecting appropriate search keywords in the future. Ultimately, more efficient and accurate information searches can be made possible, taking into account the search habits and individual attributes of the searcher as appropriate.

In addition, in the search support method of this embodiment, when the information processing device identifies the search history, it may identify, based on the attribute information and the search history, search histories by other searchers that include the searcher attribute included in the search request as an attribute, which use the identified vocabulary as a search keyword and show an occurrence frequency above a certain standard.

In addition, in the search support method of this embodiment, the information processing device may store, in the storage device, information on the name of a product or service, variations of the name, and higher or lower conceptual names of the name as the dictionary information, and when identifying the vocabulary, may compare the search keyword included in the search request with the dictionary information, and identify as the vocabulary the name of a product or service that matches the search keyword in at least one of the name, the variations, and the conceptual name.

In addition, in the search support method of this embodiment, the information processing device may store, in the storage device, information on at least one of the age, sex, location, purchasing tendency of a specified product or service, and store usage tendency of each of the searchers as the attribute information, and when identifying the search history, identify search histories by other searchers that contain, as attributes, searcher attributes that are at least one of the age, sex, location, purchasing tendency of a specified product or service, and store usage tendency of the searcher included in the search request, and that use the identified vocabulary as a search keyword.

In addition, in the search support method of this embodiment, the information processing device may execute a search using an existing search engine based only on the search keyword, and when responding, may respond with information on the search results indicated by the search history together with the results of the search.

1 Network 10 Search support system 100 Search support device 101 Storage device 102 Program 1021 Search engine 103 Memory 104 Computing device 105 Communication device 125 Attribute information DB
126 Search History DB
127 Dictionary DB
128 Habit Information DB
200 User terminal

Claims (10)

A storage device for storing attribute information and search history of each searcher who performed an information search, and dictionary information related to fields that may be searched;
a computing device that executes, when a new search request is received from a terminal, a process of comparing a search keyword included in the search request with the dictionary information and identifying vocabulary that can be regarded as being the same as or synonymous with the search keyword, a process of identifying search histories by other searchers that include the searcher attribute included in the search request as an attribute based on the attribute information and the search history, which use the identified vocabulary as a search keyword, and a process of responding to the terminal with information on the search results indicated by the identified search history;
A search support device comprising: The computing device includes:
When identifying the search history, based on the attribute information and the search history, search histories by other searchers that include the searcher attribute included in the search request as an attribute, in which the identified vocabulary is used as a search keyword and that show an occurrence frequency above a certain standard are identified.
2. The search support device according to claim 1. The storage device includes:
The dictionary information includes information on the name of a product or service, variations of the name, and higher or lower concept names of the name,
The computing device includes:
When identifying the vocabulary, a search keyword included in the search request is compared with the dictionary information, and names of products or services that match the search keyword in at least one of the name, the pronunciation, and the concept name are identified as the vocabulary.
3. The search support device according to claim 2. The storage device includes:
As the attribute information, information regarding at least one of the age, sex, location, purchasing tendency of a predetermined product or service, and store usage tendency of each of the searchers is stored;
The computing device includes:
When identifying the search history, the search history is identified by identifying search histories by other searchers that contain, as attributes, searcher attributes, which are at least any one of the searcher's age, sex, location, purchase tendency of a predetermined product or service, and store usage tendency, included in the search request, and that use the identified vocabulary as a search keyword.
4. The search support device according to claim 3. The computing device includes:
A search is executed using an existing search engine based only on the search keyword, and when responding, information on the search results indicated by the search history is responded to together with the results of the search.
2. The search support device according to claim 1. An information processing device,
A storage device is provided for storing attribute information and search history of each searcher who has performed an information search, and dictionary information relating to fields that may be searched,
When a new search request is received from a terminal, the system executes a process of comparing a search keyword included in the search request with the dictionary information and identifying vocabulary that can be regarded as being the same as or synonymous with the search keyword, a process of identifying search histories by other searchers that include the searcher attribute included in the search request as an attribute, based on the attribute information and the search history, which use the identified vocabulary as a search keyword, and a process of responding to the terminal with information on the search results indicated by the identified search history.
A search support method comprising: The information processing device,
When identifying the search history, based on the attribute information and the search history, from among search histories by other searchers that include the searcher attribute included in the search request as an attribute, those that use the identified vocabulary as a search keyword and show an occurrence frequency above a certain standard are identified.
7. The search support method according to claim 6. The information processing device,
In the storage device, the dictionary information is stored as information on the name of the product or service, the variations of the name, and the higher or lower conceptual names of the name,
When identifying the vocabulary, a search keyword included in the search request is compared with the dictionary information, and names of products or services that match the search keyword in at least one of the name, the pronunciation, and the concept name are identified as the vocabulary.
8. The search support method according to claim 7. The information processing device,
The storage device stores, as the attribute information, information regarding at least one of the age, sex, location, purchase tendency of a predetermined product or service, and store usage tendency of each of the searchers,
When identifying the search history, the search history by other searchers that includes, as an attribute, at least one of the searcher's age, sex, location, purchase tendency of a predetermined product or service, and store usage tendency, included in the search request, is identified as a search history that uses the identified vocabulary as a search keyword.
9. The search support method according to claim 8. The information processing device,
execute a search using an existing search engine based only on the search keyword, and, when responding, respond with information on the search results indicated by the search history together with the results of the search;
7. The search support method according to claim 6.

Images