CN108614832B - Method and device for realizing user personalized commodity search - Google Patents
Method and device for realizing user personalized commodity search Download PDFInfo
- Publication number
- CN108614832B CN108614832B CN201611142235.XA CN201611142235A CN108614832B CN 108614832 B CN108614832 B CN 108614832B CN 201611142235 A CN201611142235 A CN 201611142235A CN 108614832 B CN108614832 B CN 108614832B
- Authority
- CN
- China
- Prior art keywords
- commodity
- personalized
- user
- score
- feature data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000001914 filtration Methods 0.000 claims abstract description 17
- 238000000605 extraction Methods 0.000 claims description 10
- 238000012163 sequencing technique Methods 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000011144 upstream manufacturing Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 2
- 238000003672 processing method Methods 0.000 claims 1
- 230000006399 behavior Effects 0.000 description 28
- 238000010586 diagram Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to the technical field of internet, in particular to a method and a device for realizing user personalized commodity search, which enable search results to be different from person to person so as to improve the matching degree between the search intention of a user and a displayed commodity result. The searching method comprises the following steps: searching commodities according to search words input by a user to obtain preliminarily searched commodity information; filtering commodity characteristics in the user personalized characteristic data according to search words input by the user; according to the filtered user personalized feature data, extracting commodity information which accords with the user personalized feature data from the preliminarily searched commodity information to serve as the commodity information to be pushed; and sorting the commodity information to be pushed, and pushing the sorted commodity information to the user.
Description
Technical Field
The invention relates to the technical field of internet, in particular to a method and a device for realizing user personalized commodity search.
Background
Conventional search engines always present the same search results for the same search term. However, in reality, due to different search skills, even for the same search term, the demands of different types of users for the same commodity are different, that is, in many cases, users with different demands (such as similar or similar commodities) use the same search term. Therefore, the search model of the "universal" (one-size-fits-all) reduces the relevance of the search results to some extent. The new generation of search engines needs to take the preference and the current background of the user into consideration, deeply understand the search intention of the user, provide personalized search service, and enable the user to search for the psychographic goods more conveniently and more quickly. How to deeply understand the search intentions of different users and show the users with commodities meeting the search intentions is a technical problem faced by those skilled in the art.
Disclosure of Invention
The invention aims to provide a method and a device for realizing user personalized commodity search, so that search results are different from person to person, and the matching degree between the search intention of a user and a displayed commodity result is improved.
In order to achieve the above purpose, the embodiment of the present invention adopts the following technical solutions:
in a first aspect, an embodiment of the present invention provides a method for implementing a search for a user-customized product, where the method includes:
searching commodities according to search words input by a user to obtain preliminarily searched commodity information;
filtering commodity characteristics in the user personalized characteristic data according to search words input by the user;
according to the filtered user personalized feature data, extracting commodity information which accords with the user personalized feature data from the preliminarily searched commodity information to serve as the commodity information to be pushed;
and sorting the commodity information to be pushed, and pushing the sorted commodity information to the user.
With reference to the first aspect, as a first possible implementation manner, the filtering, according to a search term input by a user, a commodity feature in user personalized feature data includes:
acquiring a search word input by a user, wherein the search word comprises commodity characteristic data;
and calling the personalized feature data of the user, comparing the commodity feature data in the search word with the personalized feature data of the user, if the commodity feature data in the search word are different from the personalized feature data of the user, keeping the commodity feature data in the search word, and filtering the corresponding commodity feature data in the personalized feature data.
With reference to the first aspect or the first possible implementation manner, as a second possible implementation manner, the personalized feature data of the user is generated by the following method:
extracting commodity attributes according to behavior logs of users in the e-commerce websites;
calculating a commodity attribute feature score according to the behavior weight;
and generating personalized feature data of the user according to the user information and the calculated commodity attribute feature scores.
With reference to the second possible implementation manner of the first aspect, as a third possible implementation manner, the personalized feature data of the user is represented by a vector composed of commodity attribute feature scores.
With reference to the first aspect, as a fourth possible implementation manner, the sorting the commodity information to be pushed specifically includes:
extracting the commodity attribute of each commodity in the commodity information to be pushed, and extracting the non-personalized sequencing score corresponding to the commodity attribute;
acquiring personalized scores of commodity attributes;
calculating the personalized sorting score of the commodity according to the extracted non-personalized sorting score of the commodity attribute and the obtained personalized score of the commodity attribute;
and sorting the commodities according to the personalized sorting scores of the commodities.
With reference to the fourth possible implementation manner of the first aspect, as a fifth possible implementation manner, the calculating a personalized ranking score of a commodity specifically includes:
multiplying the extracted non-personalized sorting score of the commodity attribute by the non-personalized sorting score weight of the commodity attribute to obtain a first parameter;
multiplying the personalized score of the commodity attribute by the personalized score weight of the commodity attribute to obtain a second parameter;
adding the first parameter and the second parameter to obtain an individualized sequencing score of the commodity attribute;
and carrying out normalization processing on the personalized ranking score of each commodity attribute to obtain the personalized ranking score of the commodity.
In a second aspect, an embodiment of the present invention further provides an apparatus for implementing a search for a user-customized product, where the apparatus includes:
a search module: the system comprises a commodity searching module, a commodity searching module and a commodity searching module, wherein the commodity searching module is used for searching commodities according to search words input by a user to obtain initial searched commodity information;
a filtering module: the system comprises a database, a commodity characteristic database and a commodity characteristic database, wherein the database is used for storing commodity characteristics of a user;
an extraction module: the system comprises a database, a database server and a database, wherein the database is used for storing the information of the commodity to be pushed;
a sorting module: the system is used for sorting the commodity information to be pushed;
a pushing module: and the system is used for pushing the sorted commodity information to the user.
With reference to the second aspect, as a first possible implementation manner, the filtering module includes:
obtaining a submodule: the system comprises a search engine, a database and a database, wherein the search engine is used for acquiring search terms input by a user, and the search terms comprise commodity characteristic data;
a comparison submodule: the system is used for calling the personalized feature data of the user, comparing the commodity feature data in the search word with the personalized feature data of the user, if the commodity feature data in the search word are different from the personalized feature data of the user, keeping the commodity feature data in the search word, and filtering the corresponding commodity feature data in the personalized feature data.
With reference to the second aspect or the first possible implementation manner of the second aspect, as a second possible implementation manner, the personalized feature data of the user is generated by:
an extraction unit: the system is used for extracting commodity attributes according to behavior logs of users in an electronic commerce website;
a calculation unit: the commodity attribute feature score extracted by the extraction unit is calculated according to the behavior weight;
a generation unit: and the personalized feature data of the user is generated according to the user information and the commodity attribute feature score calculated by the calculating unit.
With reference to the second possible implementation manner of the second aspect, as a third possible implementation manner, the personalized feature data of the user is represented by a vector composed of commodity attribute feature scores.
With reference to the second aspect, as a fourth possible implementation manner, the sorting module specifically includes:
extracting a submodule: the system comprises a data processing module, a data processing module and a display module, wherein the data processing module is used for extracting commodity attributes of each commodity in commodity information to be pushed and extracting non-personalized sequencing scores corresponding to the commodity attributes;
obtaining a submodule: the system comprises a database, a database and a database, wherein the database is used for storing commodity attributes;
a calculation submodule: the personalized sorting score of the commodity is calculated according to the extracted non-personalized sorting score of the commodity attribute and the obtained personalized score of the commodity attribute;
a sorting submodule: and sorting the commodities according to the personalized sorting scores of the commodities.
With reference to the fourth possible implementation manner of the second aspect, as a fifth possible implementation manner, the calculating sub-module specifically includes:
a first acquisition unit: the non-personalized sorting score of the extracted commodity attribute is multiplied by the non-personalized sorting score weight of the commodity attribute to obtain a first parameter;
a second acquisition unit: multiplying the personalized score of the commodity attribute by the personalized score weight of the commodity attribute to obtain a second parameter;
a third acquisition unit: the system comprises a first parameter, a second parameter and a third parameter, wherein the first parameter and the second parameter are added to obtain an individualized sequencing score of the commodity attribute;
a fourth acquisition unit: and the system is used for normalizing the personalized ranking score of each commodity attribute to obtain the personalized ranking score of the commodity.
Compared with the prior art, the method and the device for realizing the user personalized commodity search can improve the matching degree between the search intention of the user and the displayed commodity result. Existing search engines always present the same search results for the same search term. According to the implementation method of the embodiment, the commodity information which accords with the personalized feature data of the user is extracted from the preliminarily searched commodity information through the personalized feature data of the user and is used as the commodity information to be pushed, then the commodity information to be pushed is sequenced, and the sequenced commodity information is pushed to the user. In the method, the personalized feature data of the user is different according to different users. Therefore, even if the same initially searched commodity information is obtained according to the same search word input by the user, different commodity information to be pushed is generated according to the personalized feature data of the user, so that the commodity information to be pushed conforms to the search intention of each user.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a block flow diagram of a method of an embodiment of the invention;
FIG. 2 is an information interaction flow diagram of an embodiment of the present invention;
fig. 3 is a block diagram of a system architecture of an embodiment of the present invention.
Detailed Description
The technical solution of the embodiment of the present invention will be described in detail below with reference to the accompanying drawings.
In order to make the technical solutions of the present invention better understood, the present invention will be described in further detail with reference to the accompanying drawings and specific embodiments. Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention, and are not to be construed as limiting the present invention. As used herein, the singular forms "a", "an", "the" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Further, "connected" or "coupled" as used herein may include wirelessly connected or coupled. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items. It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
As shown in fig. 1, an embodiment of the present invention provides a method for implementing a search for a user-customized product, including:
s10, according to the search words input by the user, carrying out commodity search to obtain the commodity information of the initial search;
s20, according to the search words input by the user, filtering the commodity characteristics in the user personalized characteristic data;
s30, according to the filtered user personalized feature data, extracting commodity information which accords with the user personalized feature data from the preliminarily searched commodity information to serve as the commodity information to be pushed;
s40, sorting the commodity information to be pushed, and pushing the sorted commodity information to the user.
By adopting the method of the embodiment, different search results can be displayed to different users. Different search results may be presented even if different users enter the same search terms. The method uses the personalized feature data of the user. The personalized feature data for different users is different. And extracting the commodity information which accords with the personalized feature data of the user from the preliminarily searched commodity information to serve as the commodity information to be pushed. Even if the initially searched commodity information is the same, the generated commodity information to be pushed is different due to different personalized feature data of the user. This enables search results to be varied from person to person, i.e. personalized searches.
In the foregoing embodiment, in step S20, the filtering the commodity features in the user personalized feature data according to the search term input by the user includes:
acquiring a search word input by a user, wherein the search word comprises commodity characteristic data;
and calling the personalized feature data of the user, comparing the commodity feature data in the search word with the personalized feature data of the user, if the commodity feature data in the search word are different from the personalized feature data of the user, keeping the commodity feature data in the search word, and filtering the corresponding commodity feature data in the personalized feature data.
When the user clearly defines the commodity characteristics in the search words, if the personalized data of the user contains the characteristics, the characteristic data is filtered from the currently extracted personalized data, namely, the characteristic is not personalized, and only the personalized characteristic data which does not appear in the search words is reserved. That is, when the user explicitly defines a characteristic of the good in the search term, the characteristic is not personalized. For example, the user searches for "black iphone6 s". Although the past history of the user shows that the user likes a white mobile phone, the personalization system still shields the personalization of the color of the mobile phone and only displays a black mobile phone as a search result.
In the above embodiments, the personalized feature data of the user is product attribute information based on user preferences. The personalized feature data of the user is established by the following method:
s201, extracting commodity attributes according to behavior logs of users in an electronic commerce website;
in this step, the attributes of the goods in which the user is interested are extracted by analyzing the user's behavior log, such as purchase, click, search history, etc., at the e-commerce site. The commodity attributes include commodity color, model, brand, style, and the like. Since the interests and behaviors of different users are different, the extracted commodity attributes are different inevitably by analyzing the behavior logs of the users on the e-commerce websites. In this way, different product attributes are extracted for different users. The user's preferences for color, function, size, etc. merchandise attributes are established for each user. In this step, the behavior log of the user in the e-commerce website in the recent period of time, for example, the behavior log in three months, may be analyzed.
S202, calculating a commodity attribute feature score according to the behavior weight;
different behaviors have different effects on the commodity attributes. For example, purchasing behavior may be more reflective of the user's search intent than click-to-view behavior. Thus, the weight of the purchasing behavior is greater than the weight of the clicking behavior. For another example, in the search history, a plurality of searches can reflect the purchasing intention of the user more than a plurality of searches. Thus, the weight of the act of searching multiple times is greater than the weight of the act of searching once. Generally, the weight of the purchasing behavior is greater than the weight of the clicking behavior, which is greater than the weight of the searching behavior. The specific value of the weight can be set according to specific situations. And calculating the commodity attribute feature score according to the behavior weight. The higher the product attribute feature score is, the greater the user's intention to search for a purchase is. In this step, the behavior log of the user in the e-commerce website in the recent period of time, for example, three months, may be analyzed to analyze each behavior and obtain the behavior weight.
And S203, generating personalized feature data of the user according to the user information and the commodity attribute feature score calculated in S202.
The user information refers to information registered by the user on the e-commerce website, and includes basic information such as age, gender, residence and the like. The personalized feature data of the user is generated according to the user information and the commodity attribute feature score calculated in S202. The personalized feature data of each user is represented by a vector consisting of feature scores of the commodity attributes. The dimensions of the vector include user information and preferred commodity characteristics such as age, color. The value stored for each dimension is the user's personalized score for that dimension.
And personalized feature data of the user is set according to the user personalization. Each user generates respective personalized feature data. The personalized feature data is different from user to user. The personalized feature data of the user generated by this step may be stored in a database, such as the REDIS database. The personalized feature data of the user can be generated by an offline module or an online module. And the personalized feature data of the user is called by the front end.
As a preferred example, in step S40, sorting the commodity information to be pushed includes:
s401, the commodity attribute of each commodity in the to-be-pushed commodity information is extracted, and the non-personalized ranking score corresponding to the commodity attribute is extracted. The non-personalized ranking score of the commodity attribute refers to the ranking score of each commodity of a common search system. The score is derived from an upstream ranking system, including consideration of relevance of the search term, inventory status, and the like.
S402, acquiring personalized scores of commodity attributes;
and if the commodity attribute is matched with the commodity attribute in the personalized feature data of the user, taking the score corresponding to the commodity attribute in the personalized feature data as the personalized score of the commodity attribute. And if the commodity attribute does not match with the commodity attribute in the personalized feature data of the user, the personalized score of the commodity attribute is regarded as 0.
S403, calculating the personalized ranking score of the commodity according to the extracted non-personalized ranking score of the commodity attribute and the obtained personalized score of the commodity attribute;
as a preferred example, the calculating of the personalized ranking score of the commodity specifically includes:
s4031, multiplying the extracted non-personalized ranking score of the commodity attribute by the non-personalized ranking score weight of the commodity attribute to obtain a first parameter;
s4032 the personalized score of the commodity attribute is multiplied by the personalized score weight of the commodity attribute to obtain a second parameter;
s4033, the first parameter and the second parameter are added to obtain the personalized ranking score of the commodity attribute;
s4034 the personalized ranking score of each commodity attribute is normalized to obtain the personalized ranking score of the commodity.
In addition to the above method, other methods may be used to calculate the personalized ranking score of the good, for example, adding the non-personalized ranking score of the extracted good attribute to the personalized score of the good attribute. By adopting the optimal selection method, the weights of different scores are comprehensively considered, and the result is normalized, so that the personalized ranking score of the commodity is more accurate.
S404, the commodities are sorted according to the personalized sorting scores of the commodities calculated in the step S403. Preferably, the items are ranked from high to low according to their personalized ranking scores.
In the sorting method, personalized sorting is based on search word relevance in common sorting. If the original common sorting results are not related, the personalized sorting cannot help the user to find a satisfactory product. For example, when the user searches for "black iPhone 7", if the results displayed by the original common ranking system are all "iPhone 6". Even if the personalized search can bring forward the black iPhone result liked by the user, the user cannot find the product desired by the user.
According to the sorting method, the commodities are sorted according to the personalized sorting scores of the commodities. The personalized ranking score of the commodity contains the parameter of the personalized score of the commodity attribute. The personalized score of the commodity attribute is obtained from the personalized feature data of the user. Thus, the personalized feature data of different users are different, and the personalized scores of the corresponding commodity attributes are also different. Therefore, the finally obtained personalized ranking score of the commodity reflects the preference of the user. The higher the personalized ranking score of the good, the greater the user's buying intent.
The method of the embodiment deeply understands the user requirements, improves the correlation between the search results and the user preferences, can help the user to further clarify or narrow the search target and range, and shortens the user search time. When the commodity information pushed to the user accords with the search intention, the page click rate can be improved, and therefore the click-to-purchase conversion rate is improved.
The method of the embodiment also improves the probability of pushing new commodity information to the user. In the method, the new commodity can be matched in a personalized way. The method of the embodiment establishes the association between the user and the commodity characteristics through the historical records of the user. After the characteristics of the new commodities are extracted, the sequence of the new commodities can be improved as long as the characteristics are consistent with the preference of the user, and the new commodities are displayed to the user. In the prior art, the commodity click times are often used as important sorting parameters, so that new commodities cannot be pushed to users due to the fact that the click times are few.
The method of the above embodiment is a personalized search engine based on e-commerce, not a personalized search engine based on web documents. Currently, a commonly used search engine based on web documents, such as a common search engine of google, Baidu, etc., mainly provides "answers" to other web documents for user search terms. The method of the embodiment mainly solves the problem of what kind of goods the user purchases at the e-commerce website. The method of the embodiment establishes the connection between the user and the commodity information in a personalized way, so that the user can quickly and conveniently buy favorite commodities on an e-commerce website. While personalization of a general search engine aims to understand a user's search terms and provide a user's desired web page result.
In the method of the above embodiment, the user preference is found by learning the history of the user and extracting the user feature data. And adjusting the sorting result to enable the product most relevant to the user to be sorted more ahead. The system can better understand the behavior characteristics of the user and provide new resources for the search service, so that a proper search result is provided for the user, the conformity degree between the search result and the search intention of the user is improved, and the search experience of the user is improved.
In the method of the above embodiment, the information pushed to the user is commodity data. And the commodity data are obtained according to the search words input by the user and the personalized feature data of the user. The method of the embodiment is to optimize the search result in the process of searching for the commodity by the user, and the feedback to the user is the specific commodity purchasing page (or link). According to the method, the webpage of the final search result is reordered by extracting, clustering and processing the user history record and matching the online product information, so that the user can find the product desired by the user in the first few items of the webpage. According to the method, through personalized search, the commodity information pushed to the user is necessarily the commodity information searched by the user, for example, the search result for searching the television is limited in the television products, but the search result can be optimized according to the favorite of the user for the brand or the function of the television.
As shown in fig. 2, the method of the embodiment of the present invention may be implemented by a client, a foreground, and a background. The client is used for inputting search terms by the user and displaying a final search result to the user by the foreground. The foreground is used for searching commodity information according to the search words input by the user. The background is used for generating personalized feature data of the user, and specifically comprises the steps of extracting commodity attributes according to a behavior log of the user; calculating a commodity attribute feature score according to the behavior weight; and generating personalized feature data of the user according to the user information and the calculated commodity attribute feature scores. When the foreground searches the commodity information, the personalized feature data of the user generated by the background needs to be called, the commodity information which is in accordance with the personalized feature data of the user is extracted from the preliminarily searched commodity information and is used as the commodity information to be pushed, then the commodity information to be pushed is sequenced, and finally the sequenced commodity information is pushed to the client and displayed to the user.
As shown in fig. 3, an embodiment of the present invention further provides an apparatus for implementing a search for a user-customized product, including:
a search module: the system comprises a commodity searching module, a commodity searching module and a commodity searching module, wherein the commodity searching module is used for searching commodities according to search words input by a user to obtain initial searched commodity information;
a filtering module: the system comprises a database, a commodity characteristic database and a commodity characteristic database, wherein the database is used for storing commodity characteristics of a user;
an extraction module: the system comprises a database, a database server and a database, wherein the database is used for storing the information of the commodity to be pushed;
a sorting module: the system is used for sorting the commodity information to be pushed;
a pushing module: and the system is used for pushing the sorted commodity information to the user.
In the above apparatus, as a preferred example, the filter module includes:
obtaining a submodule: the system comprises a search engine, a database and a database, wherein the search engine is used for acquiring search terms input by a user, and the search terms comprise commodity characteristic data;
a comparison submodule: the system is used for calling the personalized feature data of the user, comparing the commodity feature data in the search word with the personalized feature data of the user, if the commodity feature data in the search word are different from the personalized feature data of the user, keeping the commodity feature data in the search word, and filtering the corresponding commodity feature data in the personalized feature data.
In the above apparatus, the user personalized feature data may be generated by an offline module, and specifically includes the following units:
an extraction unit: the system is used for extracting commodity attributes according to behavior logs of users in an electronic commerce website;
a calculation unit: the commodity attribute feature score extracted by the extraction unit is calculated according to the behavior weight;
a generation unit: and the personalized feature data of the user is generated according to the user information and the commodity attribute feature score calculated by the calculating unit.
And generating personalized feature data of the user through the extraction unit, the calculation unit and the generation unit. The personalized feature data of the user can be represented by a vector composed of commodity attribute feature scores.
In the above apparatus, preferably, the sorting module specifically includes:
extracting a submodule: the system comprises a data processing module, a data processing module and a display module, wherein the data processing module is used for extracting commodity attributes of each commodity in commodity information to be pushed and extracting non-personalized sequencing scores corresponding to the commodity attributes;
obtaining a submodule: the system comprises a database, a database and a database, wherein the database is used for storing commodity attributes;
a calculation submodule: the personalized sorting score of the commodity is calculated according to the extracted non-personalized sorting score of the commodity attribute and the obtained personalized score of the commodity attribute;
a sorting submodule: and sorting the commodities according to the personalized sorting scores of the commodities.
Wherein, the calculation submodule specifically comprises:
a first acquisition unit: the non-personalized sorting score of the extracted commodity attribute is multiplied by the non-personalized sorting score weight of the commodity attribute to obtain a first parameter;
a second acquisition unit: multiplying the personalized score of the commodity attribute by the personalized score weight of the commodity attribute to obtain a second parameter;
a third acquisition unit: the system comprises a first parameter, a second parameter and a third parameter, wherein the first parameter and the second parameter are added to obtain an individualized sequencing score of the commodity attribute;
a fourth acquisition unit: and the system is used for normalizing the personalized ranking score of each commodity attribute to obtain the personalized ranking score of the commodity.
By adopting the device of the embodiment, different search results can be displayed to different users. Different search results may be presented even if different users enter the same search terms. The device employs personalized feature data of the user. The personalized feature data for different users is different. And extracting the commodity information which accords with the personalized feature data of the user from the preliminarily searched commodity information to serve as the commodity information to be pushed. Even if the initially searched commodity information is the same, the generated commodity information to be pushed is different due to different personalized feature data of the user. This enables search results to be varied from person to person, i.e. personalized searches.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the apparatus embodiment, since it is substantially similar to the method embodiment, it is relatively simple to describe, and reference may be made to some descriptions of the method embodiment for relevant points.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
The above description is only for the specific embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. A method for realizing user personalized commodity search is characterized by comprising the following steps:
searching commodities according to search words input by a user to obtain preliminarily searched commodity information;
filtering commodity characteristics in the user personalized characteristic data according to search words input by the user;
according to the filtered user personalized feature data, extracting commodity information which accords with the user personalized feature data from the preliminarily searched commodity information to serve as the commodity information to be pushed;
sorting the commodity information to be pushed, and pushing the sorted commodity information to a user;
the sorting of the commodity information to be pushed specifically comprises:
extracting the commodity attribute of each commodity in the commodity information to be pushed, and extracting the non-personalized sequencing score corresponding to the commodity attribute; the non-personalized ranking score of the commodity attribute refers to the ranking score of each commodity of a common search system; the ranking score is from an upstream ranking system, including relevance and inventory condition considerations for the search term;
acquiring personalized scores of commodity attributes; the method specifically comprises the following steps: if the commodity attribute is matched with the commodity attribute in the personalized feature data of the user, taking the corresponding score of the commodity attribute in the personalized feature data as the personalized score of the commodity attribute; if the commodity attribute is not matched with the commodity attribute in the personalized feature data of the user, the personalized score of the commodity attribute is regarded as 0;
calculating the personalized sorting score of the commodity according to the extracted non-personalized sorting score of the commodity attribute and the obtained personalized score of the commodity attribute;
and sorting the commodities according to the personalized sorting scores of the commodities.
2. The method according to claim 1, wherein the filtering the commodity features in the user personalized feature data according to the search terms input by the user comprises:
acquiring a search word input by a user, wherein the search word comprises commodity characteristic data;
and calling the personalized feature data of the user, comparing the commodity feature data in the search word with the personalized feature data of the user, if the commodity feature data in the search word are different from the personalized feature data of the user, keeping the commodity feature data in the search word, and filtering the corresponding commodity feature data in the personalized feature data.
3. The method according to claim 1 or 2, characterized in that the personalized feature data of the user is generated by the following method:
extracting commodity attributes according to behavior logs of users in the e-commerce websites;
calculating a commodity attribute feature score according to the behavior weight;
and generating personalized feature data of the user according to the user information and the calculated commodity attribute feature scores.
4. The method according to claim 3, wherein the personalized feature data of the user is represented by a vector composed of commodity attribute feature scores.
5. The method according to claim 1, wherein the calculating the personalized ranking score of the commodity specifically comprises:
multiplying the extracted non-personalized sorting score of the commodity attribute by the non-personalized sorting score weight of the commodity attribute to obtain a first parameter;
multiplying the personalized score of the commodity attribute by the personalized score weight of the commodity attribute to obtain a second parameter;
adding the first parameter and the second parameter to obtain an individualized sequencing score of the commodity attribute;
and carrying out normalization processing on the personalized ranking score of each commodity attribute to obtain the personalized ranking score of the commodity.
6. An apparatus for implementing a search for a user-customized article, the apparatus comprising:
a search module: the system comprises a commodity searching module, a commodity searching module and a commodity searching module, wherein the commodity searching module is used for searching commodities according to search words input by a user to obtain initial searched commodity information;
a filtering module: the system comprises a database, a commodity characteristic database and a commodity characteristic database, wherein the database is used for storing commodity characteristics of a user;
an extraction module: the system comprises a database, a database server and a database, wherein the database is used for storing the information of the commodity to be pushed;
a sorting module: the system is used for sorting the commodity information to be pushed;
a pushing module: the system is used for pushing the sorted commodity information to a user;
the sorting module specifically comprises:
extracting a submodule: the system comprises a data processing module, a data processing module and a display module, wherein the data processing module is used for extracting commodity attributes of each commodity in commodity information to be pushed and extracting non-personalized sequencing scores corresponding to the commodity attributes; the non-personalized ranking score of the commodity attribute refers to the ranking score of each commodity of a common search system; the ranking score is from an upstream ranking system, including relevance and inventory condition considerations for the search term;
obtaining a submodule: the system comprises a database, a database and a database, wherein the database is used for storing commodity attributes; the personalized feature data processing method is specifically used for taking the corresponding score of the commodity attribute in the personalized feature data as the personalized score of the commodity attribute if the commodity attribute is matched with the commodity attribute in the personalized feature data of the user; if the commodity attribute is not matched with the commodity attribute in the personalized feature data of the user, the personalized score of the commodity attribute is regarded as 0;
a calculation submodule: the personalized sorting score of the commodity is calculated according to the extracted non-personalized sorting score of the commodity attribute and the obtained personalized score of the commodity attribute;
a sorting submodule: and sorting the commodities according to the personalized sorting scores of the commodities.
7. The apparatus of claim 6, wherein the filter module comprises:
obtaining a submodule: the system comprises a search engine, a database and a database, wherein the search engine is used for acquiring search terms input by a user, and the search terms comprise commodity characteristic data;
a comparison submodule: the system is used for calling the personalized feature data of the user, comparing the commodity feature data in the search word with the personalized feature data of the user, if the commodity feature data in the search word are different from the personalized feature data of the user, keeping the commodity feature data in the search word, and filtering the corresponding commodity feature data in the personalized feature data.
8. The apparatus according to claim 6 or 7, wherein the personalized feature data of the user is generated by:
an extraction unit: the system is used for extracting commodity attributes according to behavior logs of users in an electronic commerce website;
a calculation unit: the commodity attribute feature score extracted by the extraction unit is calculated according to the behavior weight;
a generation unit: and the personalized feature data of the user is generated according to the user information and the commodity attribute feature score calculated by the calculating unit.
9. The apparatus of claim 8, wherein the personalized feature data of the user is represented by a vector composed of commodity attribute feature scores.
10. The apparatus according to claim 6, wherein the computing sub-module specifically includes:
a first acquisition unit: the non-personalized sorting score of the extracted commodity attribute is multiplied by the non-personalized sorting score weight of the commodity attribute to obtain a first parameter;
a second acquisition unit: multiplying the personalized score of the commodity attribute by the personalized score weight of the commodity attribute to obtain a second parameter;
a third acquisition unit: the system comprises a first parameter, a second parameter and a third parameter, wherein the first parameter and the second parameter are added to obtain an individualized sequencing score of the commodity attribute;
a fourth acquisition unit: and the system is used for normalizing the personalized ranking score of each commodity attribute to obtain the personalized ranking score of the commodity.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611142235.XA CN108614832B (en) | 2016-12-12 | 2016-12-12 | Method and device for realizing user personalized commodity search |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611142235.XA CN108614832B (en) | 2016-12-12 | 2016-12-12 | Method and device for realizing user personalized commodity search |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108614832A CN108614832A (en) | 2018-10-02 |
| CN108614832B true CN108614832B (en) | 2022-01-28 |
Family
ID=63657781
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611142235.XA Active CN108614832B (en) | 2016-12-12 | 2016-12-12 | Method and device for realizing user personalized commodity search |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108614832B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110083677B (en) * | 2019-05-07 | 2021-09-17 | 北京字节跳动网络技术有限公司 | Contact person searching method, device, equipment and storage medium |
| CN111028040B (en) * | 2019-06-17 | 2020-11-06 | 拉扎斯网络科技（上海）有限公司 | Object sorting method and device, readable storage medium and electronic equipment |
| CN110850997A (en) * | 2019-09-29 | 2020-02-28 | 上海萌家网络科技有限公司 | Input method and system thereof |
| CN111737418B (en) * | 2020-07-20 | 2021-05-14 | 北京每日优鲜电子商务有限公司 | Method, apparatus and storage medium for predicting relevance of search term and commodity |
| CN112102037B (en) * | 2020-09-16 | 2021-05-07 | 广州伊的家网络科技有限公司 | Live E-commerce platform commodity content intelligent pushing management system based on big data |
| CN112651839B (en) * | 2021-01-07 | 2024-07-16 | 中国农业银行股份有限公司 | Product optimization method and system |
| WO2022220738A1 (en) * | 2021-04-15 | 2022-10-20 | Grabtaxi Holdings Pte. Ltd. | Communications server apparatus, method and communications system for providing a recommendation of offerings by merchants to a user |
| CN113297511B (en) * | 2021-05-24 | 2022-09-16 | 北京三快在线科技有限公司 | Method and device for constructing merchant ranking model, server and storage medium |
| TWI826881B (en) * | 2021-12-22 | 2023-12-21 | 阿物科技股份有限公司 | System and method for awakening non-shopping consumers |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102446180A (en) * | 2010-10-09 | 2012-05-09 | 腾讯科技（深圳）有限公司 | Commodity searching method and device |
| CN104933172A (en) * | 2015-06-30 | 2015-09-23 | 百度在线网络技术（北京）有限公司 | Information pushing method and device based on user searching behavior |
| CN105446972A (en) * | 2014-06-17 | 2016-03-30 | 阿里巴巴集团控股有限公司 | Search method, device and system based on and fusing with user relation data |
-
2016
- 2016-12-12 CN CN201611142235.XA patent/CN108614832B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102446180A (en) * | 2010-10-09 | 2012-05-09 | 腾讯科技（深圳）有限公司 | Commodity searching method and device |
| CN105446972A (en) * | 2014-06-17 | 2016-03-30 | 阿里巴巴集团控股有限公司 | Search method, device and system based on and fusing with user relation data |
| CN104933172A (en) * | 2015-06-30 | 2015-09-23 | 百度在线网络技术（北京）有限公司 | Information pushing method and device based on user searching behavior |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108614832A (en) | 2018-10-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108614832B (en) | Method and device for realizing user personalized commodity search | |
| US10180979B2 (en) | System and method for generating suggestions by a search engine in response to search queries | |
| US11361239B2 (en) | Digital content classification and recommendation based upon artificial intelligence reinforcement learning | |
| US8751470B1 (en) | Context sensitive ranking | |
| US8566261B2 (en) | Interactive recommendations | |
| US20150073931A1 (en) | Feature selection for recommender systems | |
| US8380727B2 (en) | Information processing device and method, program, and recording medium | |
| US20190156395A1 (en) | System and Method for Analyzing and Searching for Features Associated with Objects | |
| KR101419504B1 (en) | System and method providing a suited shopping information by analyzing the propensity of an user | |
| JP6022056B2 (en) | Generate search results | |
| US20150379609A1 (en) | Generating recommendations for unfamiliar users by utilizing social side information | |
| US9773272B2 (en) | Recommendation engine | |
| US8725559B1 (en) | Attribute based advertisement categorization | |
| US20150254714A1 (en) | Systems and methods for keyword suggestion | |
| AU2015310494A1 (en) | Sentiment rating system and method | |
| US10366343B1 (en) | Machine learning-based literary work ranking and recommendation system | |
| US11430049B2 (en) | Communication via simulated user | |
| US11501334B2 (en) | Methods and apparatuses for selecting advertisements using semantic matching | |
| US9767204B1 (en) | Category predictions identifying a search frequency | |
| US11810148B1 (en) | Content influencer scoring system and related methods | |
| CN109829033B (en) | Data display method and terminal equipment | |
| US20250182181A1 (en) | Generating product profile recommendations and quality indicators to enhance product profiles | |
| US20150206220A1 (en) | Recommendation Strategy Portfolios | |
| CN102799593B (en) | Personalized search ranking method and system | |
| US20250044917A1 (en) | Search system for providing communications-based compatibility features |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 210000, 1-5 story, Jinshan building, 8 Shanxi Road, Nanjing, Jiangsu. Applicant after: SUNING.COM Co.,Ltd. Address before: 210042 Suning Headquarters, No. 1 Suning Avenue, Xuanwu District, Nanjing City, Jiangsu Province Applicant before: SUNING COMMERCE GROUP Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |