CN112395496A - Information recommendation method and device, electronic equipment and storage medium - Google Patents

Information recommendation method and device, electronic equipment and storage medium Download PDF

Info

Publication number
CN112395496A
CN112395496A CN202011142849.4A CN202011142849A CN112395496A CN 112395496 A CN112395496 A CN 112395496A CN 202011142849 A CN202011142849 A CN 202011142849A CN 112395496 A CN112395496 A CN 112395496A
Authority
CN
China
Prior art keywords
information
target user
candidate recommendation
recommendation information
diversity
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.)
Pending
Application number
CN202011142849.4A
Other languages
Chinese (zh)
Inventor
查强
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Zhongyuan Network Co ltd
Original Assignee
Shanghai Zhongyuan Network Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shanghai Zhongyuan Network Co ltd filed Critical Shanghai Zhongyuan Network Co ltd
Priority to CN202011142849.4A priority Critical patent/CN112395496A/en
Publication of CN112395496A publication Critical patent/CN112395496A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9535—Search customisation based on user profiles and personalisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2457—Query processing with adaptation to user needs
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2462—Approximate or statistical queries
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9538—Presentation of query results
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Computation (AREA)
  • Evolutionary Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Fuzzy Systems (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The application provides an information recommendation method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring an information recommendation request sent by a terminal of a target user; responding to the information recommendation request, and obtaining diversity preference values and candidate recommendation information of the target user, wherein the diversity preference values are obtained by browsing the log according to the historical information of the target user; screening information to be recommended from the candidate recommendation information by using the diversity preference value; and sending the information to be recommended to a terminal of a target user. According to the information recommendation method, the user behaviors are analyzed to obtain the diversity requirements of the user, the candidate recommendation information obtained after sequencing is screened according to the diversity requirements of the user, the recommendation result more conforming to the user interest is finally obtained and recommended, different diversity control can be achieved for different users, and the user experience and the online recommendation effect are optimized.

Description

Information recommendation method and device, electronic equipment and storage medium
Technical Field
The present application relates to the field of information processing technologies, and in particular, to an information recommendation method and apparatus, an electronic device, and a storage medium.
Background
The recommendation system generally displays recommendation results to the user after recalling, coarse ranking, fine ranking and strategy layers. In the top-ranking, most of the content is based on the click rate, when a certain user recommends content, the content of a certain interest point is often concentrated, although the estimated click rate is relatively high, the user is unlikely to be favored, because the interest of the user is often not unimodal. For example, a user may be interested in three categories of art, food and sports, but in the whole recommendation system, the content of the art may be better than the content of the food and sports from a priori or a posteriori, and when the system recommends the content for the user, most of the recommended content at the front part is the art content after passing through the ranking model, and the food or sports is less. Although the user has a strong variety interest, recommending only variety content tends to make the user aesthetically tired. Therefore, while meeting the accuracy of recommendation, the diversity of recommendation results is important.
In the related art, after the model is sorted and before the model is pushed to the online, a diversity control strategy is generally added, and the content is scattered through a certain rule, so that the phenomenon of content bunching of a user is avoided. At present, the existing diversity control strategy does not distinguish different diversity requirements of different users, and directly processes the diversity requirements of all users in the same way, so that the recommendation result does not meet the real requirements of the users, and the online recommendation effect is further influenced.
Disclosure of Invention
The application provides an information recommendation method and device, electronic equipment and a storage medium. According to the method, the user behaviors are analyzed to obtain the diversity requirements of the user, then the candidate recommendation information obtained after sorting is screened according to the diversity requirements of the user, and finally the recommendation result which is more in line with the user interest is obtained and recommended, different diversity control can be realized for different users, and the user experience and the online recommendation effect are optimized.
A first aspect of the present application provides an information recommendation method, including:
acquiring an information recommendation request sent by a terminal of a target user;
responding to the information recommendation request, and obtaining diversity preference values and candidate recommendation information of the target user, wherein the diversity preference values are obtained according to analysis of a historical information browsing log of the target user;
screening information to be recommended from the candidate recommendation information by using the diversity preference value;
and sending the information to be recommended to the terminal of the target user.
Optionally, screening information to be recommended from the candidate recommendation information by using the diversity preference value, including:
obtaining an evaluation score of each candidate recommendation information according to the diversity preference value of the target user, the correlation between the target user and the candidate recommendation information and the similarity between the candidate recommendation information, wherein the evaluation score represents the interest degree of the user in the candidate recommendation information;
and determining the candidate recommendation information with the preset number as the information to be recommended according to the sequence of the evaluation scores from high to low.
Optionally, obtaining an evaluation score of each candidate recommendation information according to the diversity preference value of the target user, the correlation between the target user and the candidate recommendation information, and the similarity between the candidate recommendation information, includes:
adjusting the similarity between any two different candidate recommendation information by using the diversity preference value of the target user;
and obtaining the evaluation score of each candidate recommendation information according to the correlation between the target user and the candidate recommendation information and the similarity between the adjusted candidate recommendation information.
Optionally, obtaining an evaluation score of each candidate recommendation information according to the correlation between the target user and the candidate recommendation information and the adjusted similarity between the candidate recommendation information, including:
based on a preset function and the correlation and similarity weight adjusting parameters, carrying out smooth operation on the correlation between the target user of each candidate recommendation information and the candidate recommendation information to obtain an operation result;
and obtaining the evaluation score of each candidate recommendation information according to the operation result and the similarity between the adjusted candidate recommendation information.
Optionally, adjusting the similarity between any two different pieces of candidate recommendation information by using the diversity preference value of the target user includes:
adjusting each similarity in a similarity matrix of the candidate recommendation information by using the diversity preference value of the target user;
based on a preset function and the correlation and similarity weight adjustment parameter, performing smoothing operation on the correlation between the target user of each candidate recommendation information and the candidate recommendation information to obtain an operation result, including:
based on a preset function and the correlation and similarity weight adjusting parameters, carrying out smoothing operation on the correlation between the target user of each candidate recommendation information and the candidate recommendation information to obtain a first correlation matrix and a second correlation matrix of each candidate recommendation information;
obtaining an evaluation score of each candidate recommendation information according to the operation result and the similarity between the adjusted candidate recommendation information, wherein the evaluation score comprises the following steps:
and obtaining the evaluation score of each candidate recommendation information according to the first correlation matrix and the second correlation matrix of each candidate recommendation information and the adjusted similarity matrix.
Optionally, obtaining the diversity preference value of the target user includes:
according to the historical information browsing log of the target user, counting to obtain each type of the historical information browsed by the target user and browsing frequency of each type of the historical information;
acquiring behavior divergence degrees of the target user according to each type of the historical information and browsing frequency of the historical information of each type;
and obtaining the diversity preference value of the target user according to the behavior divergence degree of the target user.
Optionally, the method further comprises:
and when the type quantity of the information browsed by the target user is less than a preset quantity, or the behavior divergence degree of the target user is greater than a preset threshold value, determining the diversity preference value of the target user as a preset value.
Optionally, the method further comprises:
acquiring historical information browsing logs of each user according to a preset period;
browsing logs according to respective historical information of each user, obtaining and storing respective diversity preference values of each user;
responding to the information recommendation request, and obtaining a diversity preference value of the target user, wherein the diversity preference value comprises the following steps:
and in response to the information recommendation request, obtaining the diversity preference value of the target user from the stored diversity preference values of the users.
A second aspect of the embodiments of the present application provides an information recommendation apparatus, including:
the first obtaining module is used for obtaining an information recommendation request sent by a terminal of a target user;
a second obtaining module, configured to obtain, in response to the information recommendation request, a diversity preference value and candidate recommendation information of the target user, where the diversity preference value is obtained according to analysis of a history information browsing log of the target user;
the screening module is used for screening information to be recommended from the candidate recommendation information by using the diversity preference value;
and the sending module is used for sending the information to be recommended to the terminal of the target user.
Optionally, the screening module comprises:
the first obtaining submodule is used for obtaining an evaluation score of each candidate recommendation information according to the diversity preference value of the target user, the correlation between the target user and the candidate recommendation information and the similarity between the candidate recommendation information, wherein the evaluation score represents the interest degree of the user in the candidate recommendation information;
and the determining submodule is used for determining the candidate recommendation information with the preset number as the information to be recommended according to the sequence of the evaluation scores from high to low.
Optionally, the first obtaining sub-module includes:
the adjusting module is used for adjusting the similarity between any two different candidate recommendation information by using the diversity preference value of the target user;
and the second obtaining submodule is used for obtaining the evaluation score of each candidate recommendation information according to the correlation between the target user and the candidate recommendation information and the similarity between the adjusted candidate recommendation information.
Optionally, the second obtaining sub-module includes:
the smoothing module is used for carrying out smoothing operation on the correlation between the target user of each candidate recommendation information and the candidate recommendation information based on a preset function and the correlation and similarity weight adjusting parameter to obtain an operation result;
and the third obtaining submodule is used for obtaining the evaluation score of each candidate recommendation information according to the operation result and the similarity between the adjusted candidate recommendation information.
Optionally, the adjusting module includes:
the adjusting submodule is used for adjusting each similarity in the similarity matrix of the candidate recommendation information by utilizing the diversity preference value of the target user;
the smoothing module includes:
a fourth obtaining submodule, configured to perform a smoothing operation on the correlation between the target user of each candidate recommendation information and the candidate recommendation information based on a preset function and the correlation and similarity weight adjustment parameter, so as to obtain a first correlation matrix and a second correlation matrix of each candidate recommendation information;
the third obtaining sub-module includes:
and the fifth obtaining submodule is used for obtaining the evaluation scores of the candidate recommendation information according to the first correlation matrix and the second correlation matrix of the candidate recommendation information and the adjusted similarity matrix.
Optionally, the second obtaining module includes:
a sixth obtaining submodule, configured to obtain, according to the historical information browsing log of the target user, statistics of each type of the historical information browsed by the target user and browsing frequency of each type of the historical information;
a seventh obtaining submodule, configured to obtain a behavior divergence degree of the target user according to each type of the historical information and browsing frequency of the historical information of each type;
and the eighth obtaining submodule is used for obtaining the diversity preference value of the target user according to the behavior divergence degree of the target user.
Optionally, the apparatus further comprises:
the determining module is used for determining the diversity preference value of the target user as a preset value when the type quantity of the information browsed by the target user is less than a preset quantity or the behavior divergence degree of the target user is greater than a preset threshold value.
Optionally, the apparatus further comprises:
the third obtaining module is used for obtaining the historical information browsing logs of each user according to a preset period;
a fourth obtaining module, configured to obtain and store a diversity preference value of each user according to the history information browsing log of each user;
the second obtaining module includes:
a ninth obtaining sub-module, configured to obtain, in response to the information recommendation request, a diversity preference value of the target user from the stored diversity preference values of the users.
A third aspect of the embodiments of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the steps in the information recommendation method according to the first aspect of the present application when executing the computer program.
A fourth aspect of the embodiments of the present application provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the steps in the information recommendation method according to the first aspect of the present application.
According to the information recommendation method provided by the application, firstly, an information recommendation request sent by a terminal of a target user is obtained; then responding to the information recommendation request, and obtaining diversity preference values and candidate recommendation information of the target user, wherein the diversity preference values are obtained by analyzing the historical information browsing logs of the target user; then, screening information to be recommended from the candidate recommendation information by using the diversity preference value; and finally, sending the information to be recommended to a terminal of a target user. According to the information recommendation method, the user behaviors are analyzed to obtain the diversity requirements of the user, the candidate recommendation information obtained after sequencing is screened according to the diversity requirements of the user, the recommendation result more conforming to the user interest is finally obtained and recommended, different diversity control can be achieved for different users, and the user experience and the online recommendation effect are optimized.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the description of the embodiments of the present application will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without inventive exercise.
FIG. 1 is a schematic diagram of an implementation environment shown in an embodiment of the present application;
FIG. 2 is a flow chart illustrating an information recommendation method according to an embodiment of the present application;
FIG. 3 is a flowchart illustrating a method for screening candidate recommendation information according to an embodiment of the present application;
FIG. 4 is a flow chart illustrating a method of obtaining an assessment score according to one embodiment of the present application;
FIG. 5 is a flow chart illustrating a method of obtaining diversity preference values according to an embodiment of the present application;
fig. 6 is a block diagram illustrating an information recommendation apparatus according to an embodiment of the present application;
fig. 7 is a schematic diagram of an electronic device according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some, but not all, embodiments of the present application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In the related art, when diversity control is performed on the candidate recommendation information, if rule control is adopted, on one hand, effective compromise between prediction accuracy and diversity cannot be achieved, flexibility is poor, on the other hand, factors considered by the rule control are limited to certain attributes of the information, and similarity between the information cannot be well measured. However, if the algorithm control is adopted, the diversity requirements of different users are not distinguished, the diversity requirements of all the users are directly treated in the same way through a fixed parameter, in practice, the diversity requirements of different users may not be the same, and therefore, the recommendation result may not meet the real requirements of the users only by the algorithm control, and the online recommendation effect is further influenced.
In order to solve the problems in the related art, the method for recommending the information is provided, the user behavior is analyzed to obtain the diversity requirements of the user, the candidate recommendation information obtained after sequencing is screened according to the diversity requirements of the user, the recommendation result more conforming to the user interest is finally obtained and recommended, different diversity control can be realized for different users, and the user experience and the online recommendation effect are optimized.
Fig. 1 is a schematic diagram of an implementation environment according to an embodiment of the present application. In fig. 1, the information recommendation platform is communicatively connected to a plurality of user terminals (including user terminal 1-user terminal N), and is configured to receive an information recommendation request sent by each user terminal and return a recommendation result in response to the information recommendation request.
The information recommendation method provided by the application is applied to the information recommendation platform shown in FIG. 1. Fig. 2 is a flowchart illustrating an information recommendation method according to an embodiment of the present application. Referring to fig. 1, the information recommendation method of the present application may include the steps of:
step S21: and obtaining an information recommendation request sent by the terminal of the target user.
In this embodiment, when a user needs to recommend information to the information recommendation platform, an information recommendation request may be sent to the information recommendation platform through the user terminal. The information requested to be recommended may be video, music, text, etc., and the type of the information is not particularly limited in the present application.
Step S22: and responding to the information recommendation request, and obtaining diversity preference values and candidate recommendation information of the target user, wherein the diversity preference values are obtained according to the analysis of the historical information browsing logs of the target user.
In this embodiment, the information recommendation platform may obtain the candidate recommendation information through steps of recalling, sorting, and the like, and how to obtain the candidate recommendation information is not specifically limited in this embodiment. The diversity preference value is a parameter for reflecting the diversity demand of the user, and is in direct proportion to the diversity demand of the user, namely when the diversity preference value is higher, the user has higher demand on the diversity, and when the diversity preference value is lower, the user has lower demand on the diversity.
In this embodiment, the diversity preference value of the target user can be obtained according to the analysis of the history information browsing log of the target user, when the history information browsed by the target user relates to multiple categories, the browsing behavior of the user is shown to be relatively divergent and the interest is wide, and when the category of the history information browsed by the user is single, the browsing behavior of the user is shown to be relatively concentrated and the interest is relatively concentrated.
Step S23: and screening information to be recommended from the candidate recommendation information by using the diversity preference value.
In this embodiment, the candidate recommendation information may be screened by using the diversity preference value to obtain the final information to be recommended.
In the present embodiment, the candidate recommendation information generally has features with two dimensions of relevance and similarity. When improving diversity on the basis of accuracy of recommendation results, relevance and similarity are two factors that must be considered. The relevance is the relevance between the user and the candidate recommendation information, and the similarity is the similarity between two different candidate recommendation information. When only the accuracy is considered, only the correlation between the user and the candidate recommendation information needs to be considered, and when only the diversity is considered, only the recommendation result needs to be scattered as much as possible according to the similarity between different candidate recommendation information. Therefore, in order to improve the recommendation effect, two dimensions of relevance and similarity should be considered comprehensively.
In specific implementation, the diversity demand degree of the target user can be determined according to the diversity preference value, and then the candidate recommendation information is screened. When the diversity preference value is higher, the relevance and the similarity can be adjusted to screen out the information to be recommended with richer varieties, and when the diversity preference value is lower, the relevance and the similarity can be adjusted to screen out the information to be recommended with more single variety.
Step S24: and sending the information to be recommended to the terminal of the target user.
In this embodiment, after the information to be recommended is obtained, the information to be recommended may be sent to the terminal where the target user is located.
Illustratively, a user A browses short videos on a short video platform X through a mobile phone, when the user A refreshes each time, the short video platform X sends an information recommendation request to a background server, the background server obtains candidate short videos through a series of steps such as recall and sorting, then a diversity preference value of the user A is obtained, if the diversity preference value is larger than a preset threshold value, the diversity requirement of the user A is larger, the background server adjusts the relevance and the similarity to obtain short videos with rich types, if the diversity preference value is not larger than the preset threshold value, the diversity requirement of the user A is smaller, the background server adjusts the relevance and the similarity to obtain short videos with concentrated types, and then the obtained short videos are sent to the mobile phone of the user to be browsed by the user.
According to the embodiment, firstly, an information recommendation request sent by a terminal of a target user is obtained; then responding to the information recommendation request, and obtaining diversity preference values and candidate recommendation information of the target user, wherein the diversity preference values are obtained by analyzing the historical information browsing logs of the target user; then, screening information to be recommended from the candidate recommendation information by using the diversity preference value; and finally, sending the information to be recommended to a terminal of a target user. According to the information recommendation method, the user behaviors are analyzed to obtain the diversity requirements of the user, the candidate recommendation information obtained after sequencing is screened according to the diversity requirements of the user, the recommendation result more conforming to the user interest is finally obtained and recommended, different diversity control can be achieved for different users, and the user experience and the online recommendation effect are optimized.
In an implementation manner, with reference to the above embodiment, the present application further provides a method for screening candidate recommendation information, as shown in fig. 3. Fig. 3 is a flowchart illustrating a method for screening candidate recommendation information according to an embodiment of the present application. Specifically, the step S23 may include:
step S231: and obtaining an evaluation score of each candidate recommendation information according to the diversity preference value of the target user, the correlation between the target user and the candidate recommendation information and the similarity between the candidate recommendation information, wherein the evaluation score represents the interest degree of the user in the candidate recommendation information.
Step S232: and determining the candidate recommendation information with the preset number as the information to be recommended according to the sequence of the evaluation scores from high to low.
In this embodiment, each candidate recommendation information may be evaluated according to a predefined rule, the diversity preference value, the correlation between the target user and the candidate recommendation information, and the similarity between the candidate recommendation information, so as to obtain a corresponding evaluation score. The evaluation score is in direct proportion to the interest degree of the user, the higher the evaluation score of the candidate recommendation information is, the more interest of the user in the candidate recommendation information is represented, and the lower the evaluation score of the candidate recommendation information is, the lower the interest degree of the user in the candidate recommendation information is represented. The predetermined rule may be any rule, and this embodiment is not particularly limited to this.
Therefore, according to the order of the evaluation scores, the candidate recommendation information with the preset number is used as the information to be recommended, so that the actual requirements of the user can be better met, and the favor of the user can be obtained.
In the embodiment, the recommendation information which is more interesting to the user can be obtained by comprehensively evaluating the diversity preference value of the target user, the correlation between the target user and the candidate recommendation information and the similarity between the candidate recommendation information and obtaining the information to be recommended according to the evaluation result, so that the recommendation effect is improved, and the user experience is optimized.
In combination with the above embodiments, in one implementation manner, the present application further provides a method for obtaining an evaluation score of each candidate recommendation information, as shown in fig. 4. Fig. 4 is a flow chart illustrating a method of obtaining an assessment score according to an embodiment of the present application. Specifically, the step S231 may include:
step S2311: and adjusting the similarity between any two different candidate recommendation information by using the diversity preference value of the target user.
Step S2312: and obtaining the evaluation score of each candidate recommendation information according to the correlation between the target user and the candidate recommendation information and the similarity between the adjusted candidate recommendation information.
In this embodiment, the similarity has a certain association with the diversity of the information, when the similarity is low, the interests of the user are relatively divergent, the types of the information are relatively rich, when the similarity is high, the interests of the user are relatively concentrated, and the types of the information are relatively single. Therefore, the similarity is adjusted according to the diversity requirements of the users, and the information to be recommended meeting the requirements of the users can be screened out.
In specific implementation, when the diversity requirement of the user is high, the similarity can be adjusted to be a low value, so that the information to be recommended with rich types can be screened out. When the diversity requirement of the user is low, the similarity can be adjusted to be a high value so as to screen out the information to be recommended with concentrated types.
In this embodiment, the similarity between the candidate recommendation information may be adjusted by using a diversity preference value, which may be any value between 0 and 1. The diversity preference value may be set to a value close to 1 when the similarity needs to be adjusted to a higher value, and may be set to a value close to 0 when the similarity needs to be adjusted to a lower value.
In the embodiment, the similarity between the candidate recommendation information is adjusted through the diversity preference value, so that the information to be recommended with abundant types or the information to be recommended with concentrated types can be screened out, and further, the targeted recommendation can be performed according to the diversity requirements of different users, the recommendation effect can be effectively improved, and the user experience is optimized.
With reference to the foregoing embodiments, in an implementation manner, the present application further provides a method for obtaining an evaluation score of each candidate recommendation information according to a correlation between a target user and the candidate recommendation information and an adjusted similarity between the candidate recommendation information. Specifically, the step S2312 may include the steps of:
based on a preset function and the correlation and similarity weight adjusting parameters, carrying out smooth operation on the correlation between the target user of each candidate recommendation information and the candidate recommendation information to obtain an operation result;
and obtaining the evaluation score of each candidate recommendation information according to the operation result and the similarity between the adjusted candidate recommendation information.
In this embodiment, since the relevance between the user and the candidate recommendation information may adopt scores from the ranking models, the final score difference of different ranking models may be relatively large, and the score directly affects the calculation result of the evaluation score of the present application, in order to facilitate tuning, the score obtained after ranking is firstly smoothed and then applied to the calculation of the evaluation score, so as to improve the accuracy of the evaluation score.
The preset function may be an exponential function, which is not limited in this embodiment. The correlation and similarity weight adjusting parameters are used for adjusting the weight between the correlation and the similarity, the smaller the correlation and similarity weight adjusting parameters are, the larger the similarity is, the similarity is shown to be influenced by the correlation, at the moment, the similarity can be adjusted on the basis of mainly taking diversity and assisting accuracy, and the richness of the type of the recommendation information is controlled by controlling the size of the similarity. The larger the adjusting parameter of the relevance and similarity weight is, the larger the influence of the relevance on the recommendation result is, and the smaller the influence of the similarity is, at the moment, the similarity can be adjusted on the basis of mainly taking accuracy and assisting diversity, and the richness of the type of the recommendation information is controlled by controlling the similarity.
In the embodiment, the correlation between the target user and the candidate recommendation information can be smoothly operated by using the preset function and the correlation and similarity weight adjustment parameter, so that the finally obtained evaluation score is more accurate, the recommendation effect is further improved, and the user experience is optimized.
In combination with the above embodiments, in one implementation, the evaluation score may be obtained by means of calculation of a matrix. Specifically, the adjusting the similarity between any two different candidate recommendation information by using the diversity preference value of the target user includes:
adjusting each similarity in a similarity matrix of the candidate recommendation information by using the diversity preference value of the target user;
based on a preset function and the correlation and similarity weight adjustment parameter, performing smoothing operation on the correlation between the target user of each candidate recommendation information and the candidate recommendation information to obtain an operation result, including:
based on a preset function and the correlation and similarity weight adjusting parameters, carrying out smoothing operation on the correlation between the target user of each candidate recommendation information and the candidate recommendation information to obtain a first correlation matrix and a second correlation matrix of each candidate recommendation information;
obtaining an evaluation score of each candidate recommendation information according to the operation result and the similarity between the adjusted candidate recommendation information, wherein the evaluation score comprises the following steps:
and obtaining the evaluation score of each candidate recommendation information according to the first correlation matrix and the second correlation matrix of each candidate recommendation information and the adjusted similarity matrix.
In an embodiment, obtaining the evaluation score of each candidate recommendation information according to the first correlation matrix and the second correlation matrix of each candidate recommendation information and the adjusted similarity matrix may include:
performing product operation on the first correlation matrix, the adjusted similarity matrix and the second correlation matrix of the target candidate recommendation information to obtain a target matrix;
determining the maximum value of the determinant of the target matrix as the evaluation score of the target candidate recommendation information.
Wherein determining a maximum value of the determinant of the target matrix as the evaluation score of the target candidate recommendation information may further include:
logarithm the determinant of the target matrix and calculating the maximum value of the logarithm;
determining the maximum value of the logarithm as the evaluation score of the target candidate recommendation information.
In this embodiment, the evaluation score of each candidate recommendation information may be obtained by the following formula:
Si,j＝(1-Wu)·<fi,fj>,i≠j
Figure BDA0002738749460000131
Figure BDA0002738749460000132
wherein R iscRepresenting a correlation between a user and the candidate recommendation information; si，jA similarity matrix representing candidate recommendation information; wuRepresenting a diversity preference value; and theta is a correlation and similarity weight adjusting parameter.
Wherein the content of the first and second substances,
Figure BDA0002738749460000141
represents a pair of RcA smoothing operation is performed.
In this embodiment, the correlation RcThe value of the determinant constructed after sorting can be taken, and the similarity S can be taken from the embedding distance between the candidate recommendation information. The weights of the two are controlled by a hyperparameter theta. Theta is a fixed value from 0 to 1 commonly used in the industry to adjust the correlation and similarity ratio of the final recommendation. In practical implementation, the requirements of different users for relevance are different, and the different requirements of different users cannot be measured by the fixed super parameter θ. Of course, the correlation RcAnd the similarity S can also be valued in other ways, which are not specifically limited in this application.
In this embodiment, det (S) can represent the diversity between two information, when S isi，jThe smaller the value of (d), the larger the det (S), the less similar the representation, i.e., the better the diversity.
In this embodiment, L is maximized when information recommendation is performedN*NWith the objective of determinism, i.e. maximizing det (L)N*N) In actual calculation, Log (det (L) is generally optimized indirectlyN*N) Both are linearly positively correlated). Since θ can represent correlation and similarityThe adjustment parameter between degrees, the smaller theta,
Figure BDA0002738749460000142
the smaller the effect, at this point Log (det (L)N*N) The more log (det (S)) affects the results. When the maximum det (L)N*N) In this case, the log (det (S)) is maximized, i.e., the recommendation results are more biased toward diversity.
WuIndicates diversity preference values due to det (S) and Si，jIs reversed, so that S can be controlled as Wu is smaller and 1-Wu is largeri，jIs a larger value, so that det (S) is smaller, i.e., the diversity is worse. Conversely, the larger Wu, the smaller 1-Wu, the S can be controlledi，jIs a small value, so that det (S) is large, i.e., diversity is better.
By the above formula, det (L) can be expressedN*N) Maximum value of (d) or Log (det (L)N*N) ) as the evaluation score of each candidate recommendation information.
According to the method for calculating the evaluation score of the candidate recommendation information, the candidate recommendation information with the larger evaluation score can be screened from the candidate recommendation information and used as the candidate recommendation information interested by the user for recommendation, the recommendation effect can be effectively improved, and the user experience is optimized.
In combination with the above embodiments, in one implementation manner, the present application further provides a method for obtaining a diversity preference value of a target user, as shown in fig. 5. Fig. 5 is a flowchart illustrating a method of obtaining diversity preference values according to an embodiment of the present application. Specifically, the step S22 may include:
step S221: and according to the historical information browsing log of the target user, counting to obtain each type of the historical information browsed by the target user and the browsing frequency of each type of the historical information.
In this embodiment, the history information browsing log of the target user may be analyzed to obtain each type of history information browsed by the target user and browsing frequency under each type. For example, when the information is a video, the various types obtained may include: entertainment, news, science and technology, food, movies, photography, etc.; when the information is news, the various types obtained may include: entertainment, science, food, movies, photography, and the like. The present embodiment does not specifically limit each type of information.
For one type of information, the user may browse multiple times, for example, when the information is a video, the user may browse multiple videos under the sub-type of entertainment, or may browse multiple videos under the sub-type of technology.
Step S222: and acquiring the behavior divergence degree of the target user according to each type of the historical information and the browsing frequency of the historical information of each type.
In this embodiment, the behavior divergence degree can be expressed by using information entropy (the information entropy can measure the divergence degree of the user behavior, and the smaller the entropy value, the lower the divergence degree, that is, the lower the requirement of the user on diversity).
In one embodiment, step S222 may include:
respectively quoting the browsing frequency of each type of historical information and the browsing frequency of all types of historical information to obtain the probability of the target user browsing each type of information;
logarithm is carried out on the probability of each type of information;
and summing and negating products of the probabilities and the corresponding logarithms in sequence to obtain the information entropy of the target user.
Specifically, the information entropy of the target user can be calculated by adopting the following formula:
Figure BDA0002738749460000161
where k represents the number of types of history information and Pi represents the probability of the user browsing the i-th type of information.
Illustratively, the types of historical information viewed by the target user include A, B, C, with the target user viewing a type A of information a number of timesThe number is 2, the number of times the B-type information is viewed is 3, the number of times the C-type information is viewed is 4, then for this time k should be 3, pi (a) ═ 2/9; pi (b) ═ 3/9; pi (c) ═ 4/9. Therefore, the temperature of the molten metal is controlled,
Figure BDA0002738749460000162
Figure BDA0002738749460000163
Figure BDA0002738749460000164
step S223: and obtaining the diversity preference value of the target user according to the behavior divergence degree of the target user.
In this embodiment, after the information entropy is calculated, the information entropy may be used as the diversity preference value of the target user. Since the behavior divergence degrees of different users are different, the diversity preference values of different users are also different.
Through the embodiment, the accumulated behavior distribution of the user to different types of data can be counted by analyzing the historical behaviors of the user, the diversity preference value of the user is calculated, then the candidate recommendation information obtained after sequencing is screened according to the diversity preference value of the user, finally, the recommendation result more conforming to the user interest is obtained and recommended, different diversity control can be realized for different users, and the user experience and the online recommendation effect are optimized.
In combination with the above embodiments, in one implementation, when obtaining the diversity preference value, the application may further include the following steps:
and when the type quantity of the information browsed by the target user is less than a preset quantity, or the behavior divergence degree of the target user is greater than a preset threshold value, determining the diversity preference value of the target user as a preset value.
In actual implementation, the diversity preference value may also be set according to a rule. For example, when the browsing frequency of a certain user is lower than a preset frequency (for example, a new user), the information entropy cannot be calculated according to a given formula, and the information entropy of the user may be set to a preset value. Or when the user behavior of a certain user is particularly divergent, which results in that the calculated information entropy is greater than a certain set value, the information entropy of the user can also be set to be a preset value.
In one embodiment, the preset value may be a median of information entropies of other users. Of course, the preset value may also be set as another manually specified value according to actual requirements, and this embodiment is not particularly limited thereto.
By the embodiment, the diversity preference value of the user with the browsing frequency lower than the preset frequency and the diversity preference value of the user with the large behavior divergence degree can be obtained, so that the information recommendation method can perform targeted recommendation according to different diversity requirements of different users.
With reference to the foregoing embodiment, in an implementation manner, the information recommendation method of the present application may further include the following steps:
acquiring historical information browsing logs of each user according to a preset period;
browsing logs according to respective historical information of each user, obtaining and storing respective diversity preference values of each user;
responding to the information recommendation request, and obtaining a diversity preference value of the target user, wherein the diversity preference value comprises the following steps:
and in response to the information recommendation request, obtaining the diversity preference value of the target user from the stored diversity preference values of the users.
In this embodiment, the information recommendation platform may periodically obtain the history information browsing logs of each user, and then obtain and store the respective diversity preference values of each user according to the respective history information browsing logs of each user. When receiving an information recommendation request of a target user, the information recommendation platform obtains diversity preference values of the target user from the pre-stored diversity preference values.
The user behaviors are analyzed to obtain the diversity requirements of the users, the candidate recommendation information obtained after sequencing is screened according to the diversity requirements of the users, the recommendation results more conforming to the user interests are finally obtained and recommended, different diversity control can be realized for different users, and the user experience and the online recommendation effect are optimized.
It should be noted that, for simplicity of description, the method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present invention is not limited by the illustrated order of acts, as some steps may occur in other orders or concurrently in accordance with the embodiments of the present invention. Further, those skilled in the art will appreciate that the embodiments described in the specification are presently preferred and that no particular act is required to implement the invention.
Based on the same inventive concept, an embodiment of the present application provides a recommendation probability estimation apparatus 600. Fig. 6 is a block diagram of an information recommendation apparatus according to an embodiment of the present application. As shown in fig. 6, the information recommendation apparatus 600 includes:
a first obtaining module 601, configured to obtain an information recommendation request sent by a terminal of a target user;
a second obtaining module 602, configured to obtain, in response to the information recommendation request, a diversity preference value and candidate recommendation information of the target user, where the diversity preference value is obtained according to analysis of a historical information browsing log of the target user;
a screening module 603, configured to screen information to be recommended from the candidate recommendation information by using the diversity preference value;
a sending module 604, configured to send the information to be recommended to the terminal of the target user.
Optionally, the screening module 603 includes:
the first obtaining submodule is used for obtaining an evaluation score of each candidate recommendation information according to the diversity preference value of the target user, the correlation between the target user and the candidate recommendation information and the similarity between the candidate recommendation information, wherein the evaluation score represents the interest degree of the user in the candidate recommendation information;
and the determining submodule is used for determining the candidate recommendation information with the preset number as the information to be recommended according to the sequence of the evaluation scores from high to low.
Optionally, the first obtaining sub-module includes:
the adjusting module is used for adjusting the similarity between any two different candidate recommendation information by using the diversity preference value of the target user;
and the second obtaining submodule is used for obtaining the evaluation score of each candidate recommendation information according to the correlation between the target user and the candidate recommendation information and the similarity between the adjusted candidate recommendation information.
Optionally, the second obtaining sub-module includes:
the smoothing module is used for carrying out smoothing operation on the correlation between the target user of each candidate recommendation information and the candidate recommendation information based on a preset function and the correlation and similarity weight adjusting parameter to obtain an operation result;
and the third obtaining submodule is used for obtaining the evaluation score of each candidate recommendation information according to the operation result and the similarity between the adjusted candidate recommendation information.
Optionally, the adjusting module includes:
the adjusting submodule is used for adjusting each similarity in the similarity matrix of the candidate recommendation information by utilizing the diversity preference value of the target user;
the smoothing module includes:
a fourth obtaining submodule, configured to perform a smoothing operation on the correlation between the target user of each candidate recommendation information and the candidate recommendation information based on a preset function and the correlation and similarity weight adjustment parameter, so as to obtain a first correlation matrix and a second correlation matrix of each candidate recommendation information;
the third obtaining sub-module includes:
and the fifth obtaining submodule is used for obtaining the evaluation scores of the candidate recommendation information according to the first correlation matrix and the second correlation matrix of the candidate recommendation information and the adjusted similarity matrix.
Optionally, the second obtaining module 602 includes:
a sixth obtaining submodule, configured to obtain, according to the historical information browsing log of the target user, statistics of each type of the historical information browsed by the target user and browsing frequency of each type of the historical information;
a seventh obtaining submodule, configured to obtain a behavior divergence degree of the target user according to each type of the historical information and browsing frequency of the historical information of each type;
and the eighth obtaining submodule is used for obtaining the diversity preference value of the target user according to the behavior divergence degree of the target user.
Optionally, the apparatus 600 further comprises:
the determining module is used for determining the diversity preference value of the target user as a preset value when the type quantity of the information browsed by the target user is less than a preset quantity or the behavior divergence degree of the target user is greater than a preset threshold value.
Optionally, the apparatus 600 further comprises:
the third obtaining module is used for obtaining the historical information browsing logs of each user according to a preset period;
a fourth obtaining module, configured to obtain and store a diversity preference value of each user according to the history information browsing log of each user;
the second obtaining module 602 includes:
a ninth obtaining sub-module, configured to obtain, in response to the information recommendation request, a diversity preference value of the target user from the stored diversity preference values of the users.
Based on the same inventive concept, another embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the steps in the method according to any of the above-mentioned embodiments of the present application.
Based on the same inventive concept, another embodiment of the present application provides an electronic device 700, as shown in fig. 7. Fig. 7 is a schematic diagram of an electronic device according to an embodiment of the present application. The electronic device comprises a memory 702, a processor 701 and a computer program stored on the memory and executable on the processor, which when executed implements the steps of the method according to any of the embodiments of the present application.
For the device embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.
The embodiments in the present specification are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.
As will be appreciated by one of skill in the art, embodiments of the present application may be provided as a method, apparatus, or computer program product. Accordingly, embodiments of the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
Embodiments of the present application are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing terminal to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing terminal to cause a series of operational steps to be performed on the computer or other programmable terminal to produce a computer implemented process such that the instructions which execute on the computer or other programmable terminal provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present application have been described, additional variations and modifications of these embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including the preferred embodiment and all such alterations and modifications as fall within the true scope of the embodiments of the application.
Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or terminal that comprises the element.
The information recommendation method, the information recommendation device, the storage medium and the electronic device provided by the application are introduced in detail, and a specific example is applied in the description to explain the principle and the implementation of the application, and the description of the embodiment is only used to help understand the method and the core idea of the application; meanwhile, for a person skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.

Claims (11)

1. An information recommendation method, comprising:
acquiring an information recommendation request sent by a terminal of a target user;
responding to the information recommendation request, and obtaining diversity preference values and candidate recommendation information of the target user, wherein the diversity preference values are obtained according to analysis of a historical information browsing log of the target user;
screening information to be recommended from the candidate recommendation information by using the diversity preference value;
and sending the information to be recommended to the terminal of the target user.
2. The method of claim 1, wherein the using the diversity preference value to screen the candidate recommendation information for information to be recommended comprises:
obtaining an evaluation score of each candidate recommendation information according to the diversity preference value of the target user, the correlation between the target user and the candidate recommendation information and the similarity between the candidate recommendation information, wherein the evaluation score represents the interest degree of the user in the candidate recommendation information;
and determining the candidate recommendation information with the preset number as the information to be recommended according to the sequence of the evaluation scores from high to low.
3. The method of claim 2, wherein obtaining the evaluation score of each candidate recommendation information according to the diversity preference value of the target user, the correlation between the target user and the candidate recommendation information, and the similarity between the candidate recommendation information comprises:
adjusting the similarity between any two different candidate recommendation information by using the diversity preference value of the target user;
and obtaining the evaluation score of each candidate recommendation information according to the correlation between the target user and the candidate recommendation information and the similarity between the adjusted candidate recommendation information.
4. The method of claim 3, wherein obtaining the evaluation score of each candidate recommendation information according to the correlation between the target user and the candidate recommendation information and the adjusted similarity between the candidate recommendation information comprises:
based on a preset function and the correlation and similarity weight adjusting parameters, carrying out smooth operation on the correlation between the target user of each candidate recommendation information and the candidate recommendation information to obtain an operation result;
and obtaining the evaluation score of each candidate recommendation information according to the operation result and the similarity between the adjusted candidate recommendation information.
5. The method of claim 4, wherein adjusting the similarity between any two different candidate recommendation information using the diversity preference value of the target user comprises:
adjusting each similarity in a similarity matrix of the candidate recommendation information by using the diversity preference value of the target user;
based on a preset function and the correlation and similarity weight adjustment parameter, performing smoothing operation on the correlation between the target user of each candidate recommendation information and the candidate recommendation information to obtain an operation result, including:
based on a preset function and the correlation and similarity weight adjusting parameters, carrying out smoothing operation on the correlation between the target user of each candidate recommendation information and the candidate recommendation information to obtain a first correlation matrix and a second correlation matrix of each candidate recommendation information;
obtaining an evaluation score of each candidate recommendation information according to the operation result and the similarity between the adjusted candidate recommendation information, wherein the evaluation score comprises the following steps:
and obtaining the evaluation score of each candidate recommendation information according to the first correlation matrix and the second correlation matrix of each candidate recommendation information and the adjusted similarity matrix.
6. The method of claim 1, wherein obtaining a diversity preference value for the target user comprises:
according to the historical information browsing log of the target user, counting to obtain each type of the historical information browsed by the target user and browsing frequency of each type of the historical information;
acquiring behavior divergence degrees of the target user according to each type of the historical information and browsing frequency of the historical information of each type;
and obtaining the diversity preference value of the target user according to the behavior divergence degree of the target user.
7. The method of claim 6, further comprising:
and when the type quantity of the information browsed by the target user is less than a preset quantity, or the behavior divergence degree of the target user is greater than a preset threshold value, determining the diversity preference value of the target user as a preset value.
8. The method of claim 1, further comprising:
acquiring historical information browsing logs of each user according to a preset period;
browsing logs according to respective historical information of each user, obtaining and storing respective diversity preference values of each user;
responding to the information recommendation request, and obtaining a diversity preference value of the target user, wherein the diversity preference value comprises the following steps:
and in response to the information recommendation request, obtaining the diversity preference value of the target user from the stored diversity preference values of the users.
9. An information recommendation apparatus, comprising:
the first obtaining module is used for obtaining an information recommendation request sent by a terminal of a target user;
a second obtaining module, configured to obtain, in response to the information recommendation request, a diversity preference value and candidate recommendation information of the target user, where the diversity preference value is obtained according to analysis of a history information browsing log of the target user;
the screening module is used for screening information to be recommended from the candidate recommendation information by using the diversity preference value;
and the sending module is used for sending the information to be recommended to the terminal of the target user.
10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing performs the steps of the information recommendation method according to any one of claims 1-8.
11. A computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, carries out the steps of the information recommendation method according to any one of claims 1-8.
CN202011142849.4A 2020-10-22 2020-10-22 Information recommendation method and device, electronic equipment and storage medium Pending CN112395496A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011142849.4A CN112395496A (en) 2020-10-22 2020-10-22 Information recommendation method and device, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011142849.4A CN112395496A (en) 2020-10-22 2020-10-22 Information recommendation method and device, electronic equipment and storage medium

Publications (1)

Publication Number Publication Date
CN112395496A true CN112395496A (en) 2021-02-23

Family

ID=74595662

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011142849.4A Pending CN112395496A (en) 2020-10-22 2020-10-22 Information recommendation method and device, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN112395496A (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112989214A (en) * 2021-05-20 2021-06-18 湖北游游智游网络科技有限公司 Tourism information display method and related equipment
CN113360758A (en) * 2021-06-08 2021-09-07 苍穹数码技术股份有限公司 Information recommendation method and device, electronic equipment and computer storage medium
CN113420214A (en) * 2021-06-23 2021-09-21 青岛海信智慧生活科技股份有限公司 Electronic transaction object recommendation method, device and equipment
CN113515696A (en) * 2021-05-17 2021-10-19 上海众源网络有限公司 Recommendation method and device, electronic equipment and storage medium
CN114722238A (en) * 2022-05-13 2022-07-08 北京达佳互联信息技术有限公司 Video recommendation method and device, electronic equipment, storage medium and program product

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103294795A (en) * 2013-05-24 2013-09-11 华东师范大学 Method for adjusting film recommending diversity by utilizing users' characters
CN108647227A (en) * 2018-03-26 2018-10-12 北京奇艺世纪科技有限公司 A kind of recommendation method and device
CN111010592A (en) * 2019-12-19 2020-04-14 上海众源网络有限公司 Video recommendation method and device, electronic equipment and storage medium
CN111291217A (en) * 2020-03-04 2020-06-16 北京奇艺世纪科技有限公司 Content recommendation method and device, electronic equipment and computer readable medium
CN111400615A (en) * 2020-03-19 2020-07-10 腾讯科技（深圳）有限公司 Resource recommendation method, device, equipment and storage medium
CN111708950A (en) * 2020-06-22 2020-09-25 腾讯科技（深圳）有限公司 Content recommendation method and device and electronic equipment

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103294795A (en) * 2013-05-24 2013-09-11 华东师范大学 Method for adjusting film recommending diversity by utilizing users' characters
CN108647227A (en) * 2018-03-26 2018-10-12 北京奇艺世纪科技有限公司 A kind of recommendation method and device
CN111010592A (en) * 2019-12-19 2020-04-14 上海众源网络有限公司 Video recommendation method and device, electronic equipment and storage medium
CN111291217A (en) * 2020-03-04 2020-06-16 北京奇艺世纪科技有限公司 Content recommendation method and device, electronic equipment and computer readable medium
CN111400615A (en) * 2020-03-19 2020-07-10 腾讯科技（深圳）有限公司 Resource recommendation method, device, equipment and storage medium
CN111708950A (en) * 2020-06-22 2020-09-25 腾讯科技（深圳）有限公司 Content recommendation method and device and electronic equipment

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113515696A (en) * 2021-05-17 2021-10-19 上海众源网络有限公司 Recommendation method and device, electronic equipment and storage medium
CN112989214A (en) * 2021-05-20 2021-06-18 湖北游游智游网络科技有限公司 Tourism information display method and related equipment
CN113360758A (en) * 2021-06-08 2021-09-07 苍穹数码技术股份有限公司 Information recommendation method and device, electronic equipment and computer storage medium
CN113420214A (en) * 2021-06-23 2021-09-21 青岛海信智慧生活科技股份有限公司 Electronic transaction object recommendation method, device and equipment
CN113420214B (en) * 2021-06-23 2023-01-10 青岛海信智慧生活科技股份有限公司 Electronic transaction object recommendation method, device and equipment
CN114722238A (en) * 2022-05-13 2022-07-08 北京达佳互联信息技术有限公司 Video recommendation method and device, electronic equipment, storage medium and program product
CN114722238B (en) * 2022-05-13 2022-09-30 北京达佳互联信息技术有限公司 Video recommendation method and device, electronic equipment, storage medium and program product

Similar Documents

Publication Publication Date Title
CN112395496A (en) Information recommendation method and device, electronic equipment and storage medium
CN109902708B (en) Recommendation model training method and related device
CN109086439B (en) Information recommendation method and device
JP4524709B2 (en) Information processing apparatus and method, and program
US10789634B2 (en) Personalized recommendation method and system, and computer-readable record medium
US10380649B2 (en) System and method for logistic matrix factorization of implicit feedback data, and application to media environments
US20170169040A1 (en) Method and electronic device for recommending video
Hariri et al. Context adaptation in interactive recommender systems
CN107341176B (en) Sample weight setting method and device and electronic equipment
KR20160064447A (en) A recommendation method for new users by using preference prediction based on collaborative filtering algorithm
CN109168047B (en) Video recommendation method and device, server and storage medium
US20190278819A1 (en) Systems and methods of providing recommendations of content items
CN109996122B (en) Video recommendation method and device, server and storage medium
CN109558514B (en) Video recommendation method, device thereof, information processing equipment and storage medium
US20180365709A1 (en) Personalized creator recommendations
US20130073618A1 (en) Information Providing System, Information Providing method, Information Providing Device, Program, And Information Storage Medium
CN107577736B (en) File recommendation method and system based on BP neural network
CN106127506B (en) recommendation method for solving cold start problem of commodity based on active learning
Adomavicius et al. Classification, ranking, and top-K stability of recommendation algorithms
US20230300417A1 (en) Video content recommendation method and apparatus, and computer device
CN112579913B (en) Video recommendation method, device, equipment and computer readable storage medium
CN113656681B (en) Object evaluation method, device, equipment and storage medium
CN111241381A (en) Information recommendation method and device, electronic equipment and computer-readable storage medium
CN112765484A (en) Short video pushing method and device, electronic equipment and storage medium
CN109636530B (en) Product determination method, product determination device, electronic equipment and computer-readable storage medium

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