CN116112710A - Information recommendation method, device and server - Google Patents
Information recommendation method, device and server Download PDFInfo
- Publication number
- CN116112710A CN116112710A CN202310134565.8A CN202310134565A CN116112710A CN 116112710 A CN116112710 A CN 116112710A CN 202310134565 A CN202310134565 A CN 202310134565A CN 116112710 A CN116112710 A CN 116112710A
- Authority
- CN
- China
- Prior art keywords
- target user
- topics
- target
- video content
- topic
- 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
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
- H04N21/251—Learning process for intelligent management, e.g. learning user preferences for recommending movies
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
- H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
- H04N21/4662—Learning process for intelligent management, e.g. learning user preferences for recommending movies characterized by learning algorithms
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
- H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
- H04N21/4667—Processing of monitored end-user data, e.g. trend analysis based on the log file of viewer selections
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
- H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
- H04N21/4668—Learning process for intelligent management, e.g. learning user preferences for recommending movies for recommending content, e.g. movies
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Computing Systems (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides an information recommendation method, an information recommendation device and a server, and relates to the technical field of computers. The information recommendation method comprises the following steps: determining a preference tag of the target user from a plurality of tags of the historical video according to the operation behavior of the target user on the historical video, wherein the historical video is the video watched by the target user in the history; determining a plurality of interesting topics of the target user from a plurality of topics corresponding to the topic labels according to the topic labels and the preference labels in the topic library; determining a target theme from the plurality of topics of interest according to video content in the plurality of topics of interest; and recommending the video content in the target theme to the terminal equipment corresponding to the target user. The preference labels of the target users and a plurality of interesting topics of the target users are automatically determined, and then the target topics are determined, so that the automatic determination of the target topics can be realized, manual screening is not needed, human resources are saved, and the topic recommendation efficiency is improved.
Description
Technical Field
The present invention relates to the field of computer technologies, and in particular, to an information recommendation method, an information recommendation device, and a server.
Background
Internet television refers to a method for transmitting multimedia audio-visual services including video, audio, graphics, text, data and the like based on the public internet and providing the public with safe, reliable, interactive and manageable multimedia audio-visual services through terminal receiving devices such as televisions, set top boxes and the like. With the popularization of internet televisions, topic recommendation is also widely applied to television products.
In the related technology, a topic recommendation mode is to generate a core concept of a topic based on operation experience and popular trend definition by an operator, then manually screen out a program composition topic conforming to the defined core concept based on the defined core concept, and deploy topic configuration to a terminal to realize topic recommendation.
However, in the related art, the programs need to be manually screened out to form the theme, which wastes unnecessary human resources and reduces the efficiency of theme recommendation.
Disclosure of Invention
The present invention has been made in view of the above-mentioned drawbacks of the related art, and an object of the present invention is to provide an information recommendation method, apparatus, server, and storage medium, which solve the above-mentioned problems occurring in the related art.
In order to achieve the above purpose, the technical scheme adopted by the embodiment of the invention is as follows:
In a first aspect, an embodiment of the present invention provides an information recommendation method, including:
determining a preference tag of a target user from a plurality of tags of a historical video according to the operation behavior of the target user on the historical video, wherein the historical video is a video which is watched by the target user in a historical manner;
determining a plurality of interesting topics of the target user from a plurality of topics corresponding to the topic tags according to the topic tags and the preference tags in the topic library;
determining a target theme from the plurality of topics of interest according to video content in the plurality of topics of interest;
and recommending the video content in the target theme to the terminal equipment corresponding to the target user.
Optionally, the determining, according to the operation behavior of the target user on the historical video, the preference tag of the target user from multiple tags of the historical video includes:
determining preference scores of the target user for a plurality of labels of each label dimension according to the operation behaviors of the target user for the historical video;
sorting the labels of each label dimension according to the preference scores of the labels of each label dimension to obtain sorted labels of each label dimension;
And determining the preference label of each label dimension from the plurality of ordered labels of each label dimension.
Optionally, the determining, according to the operation behavior of the target user for the historical video, preference scores of the target user for a plurality of tags of each tag dimension includes:
for each historical time period, obtaining an initial score of the target user for each tag in each historical time period according to the behavior type of the operation behavior, the behavior weight of the behavior type, the times of the operation behavior of the target user for each tag in a plurality of tags of each tag dimension, the times of the operation behavior of the target user for all tags in the target tag dimension corresponding to each tag, the times of the operation behavior of the target user for all tags in the target tag dimension, and the times of the operation behavior of the target user for each tag;
and accumulating the initial scores of each historical time period to obtain preference scores of the target user for a plurality of labels of each label dimension.
Optionally, the determining, according to the topic tags in the topic library and the preference tags, a plurality of topics of interest of the target user from a plurality of topics corresponding to the topic tags includes:
matching the theme labels and the preference labels to obtain interest scores of the target user for the theme labels;
sorting the topics according to the interest scores of the topics to obtain sorted topics;
and determining a plurality of interesting topics of the target user from the ordered topics.
Optionally, the matching the plurality of topic tags and the preference tag to obtain interest scores of the target user for the plurality of topics includes:
and calculating according to the preset weight corresponding to the label dimension, the matching quantity of the preference label and each theme label in the corresponding label dimension and the corresponding preset value in the label dimension to obtain interest scores of the themes.
Optionally, the determining, according to the video content in the multiple topics of interest, a target topic from the multiple topics of interest includes:
Scoring the video content in each interested subject to obtain the score of each video content in each interested subject;
determining target video content of each interested subject according to the score of each video content in each interested subject;
obtaining the score of each interested subject according to the score of the target video content of each interested subject;
and determining a target theme from the multiple themes of interest according to the score of each theme of interest.
Optionally, the scoring the video content in each topic of interest to obtain a score of each video content in each topic of interest includes:
acquiring the characteristic information of the target user and the characteristic information of the video content in each interested subject;
and processing the characteristic information of the target user and the characteristic information of the video content in each interested subject by adopting a preset video content scoring model to obtain the score of each video content in each interested subject.
Optionally, the method further comprises:
and performing model training according to the characteristic information of the sample user, the first sample video content which is exposed and requested by the sample user, and the second sample video content which is exposed and not requested by the sample user, so as to obtain the preset video content scoring model.
In a second aspect, an embodiment of the present invention further provides an information recommendation apparatus, including:
the determining module is used for determining preference tags of the target user from a plurality of tags of the historical video according to the operation behaviors of the target user on the historical video, wherein the historical video is the video which is historically watched by the target user; determining a plurality of interesting topics of the target user from a plurality of topics corresponding to the topic tags according to the topic tags and the preference tags in the topic library; determining a target theme from the plurality of topics of interest according to video content in the plurality of topics of interest;
and the recommending module is used for recommending the video content in the target theme to the terminal equipment corresponding to the target user.
Optionally, the determining module is specifically configured to determine, according to an operation behavior of the target user with respect to the historical video, preference scores of the target user with respect to a plurality of tags of each tag dimension; sorting the labels of each label dimension according to the preference scores of the labels of each label dimension to obtain sorted labels of each label dimension; and determining the preference label of each label dimension from the plurality of ordered labels of each label dimension.
Optionally, the determining module is specifically configured to obtain, for each historical time period, an initial score of the target user for each tag in the historical time period according to a behavior type of the operation behavior, a behavior weight of the behavior type, a number of times the target user performs the operation behavior for each tag in a plurality of tags in each tag dimension, a number of times the target user performs the operation behavior for all tags in a target tag dimension corresponding to the each tag, a number of times the target user performs the operation behavior for all tags in the target tag dimension, and a number of times the target user performs the operation behavior for each tag; and accumulating the initial scores of each historical time period to obtain preference scores of the target user for a plurality of labels of each label dimension.
Optionally, the determining module is specifically configured to match the plurality of topic tags with the preference tag to obtain interest scores of the target user for the plurality of topics; sorting the topics according to the interest scores of the topics to obtain sorted topics; and determining a plurality of interesting topics of the target user from the ordered topics.
Optionally, the determining module is specifically configured to calculate according to a preset weight corresponding to a tag dimension, a number of matches between the preference tag and each topic tag in the corresponding tag dimension, and a preset value corresponding to the tag dimension, so as to obtain interest scores of the multiple topics.
Optionally, the determining module is specifically configured to score video content in each topic of interest, so as to obtain a score of each video content in each topic of interest; determining target video content of each interested subject according to the score of each video content in each interested subject; obtaining the score of each interested subject according to the score of the target video content of each interested subject; and determining a target theme from the multiple themes of interest according to the score of each theme of interest.
Optionally, the determining module is specifically configured to obtain feature information of the target user and feature information of video content in each topic of interest; and processing the characteristic information of the target user and the characteristic information of the video content in each interested subject by adopting a preset video content scoring model to obtain the score of each video content in each interested subject.
Optionally, the apparatus further includes:
and the training module is used for carrying out model training according to the characteristic information of the sample user, the first sample video content which is exposed and requested by the sample user, and the second sample video content which is exposed and not requested by the sample user, so as to obtain the preset video content scoring model.
In a third aspect, an embodiment of the present invention further provides a server, including: a memory storing a computer program executable by the processor, and a processor implementing the information recommendation method according to any one of the above first aspects when the processor executes the computer program.
In a fourth aspect, an embodiment of the present invention further provides a computer readable storage medium, where a computer program is stored, where the computer program is read and executed to implement the information recommendation method according to any one of the first aspects.
The beneficial effects of the invention are as follows: the embodiment of the invention provides an information recommendation method, which comprises the following steps: determining a preference tag of the target user from a plurality of tags of the historical video according to the operation behavior of the target user on the historical video, wherein the historical video is the video watched by the target user in the history; determining a plurality of interesting topics of the target user from a plurality of topics corresponding to the topic labels according to the topic labels and the preference labels in the topic library; determining a target theme from the plurality of topics of interest according to video content in the plurality of topics of interest; and recommending the video content in the target theme to the terminal equipment corresponding to the target user. The preference labels of the target users and the multiple interested topics of the target users are automatically determined, and then the target topics are determined from the multiple interested topics according to video content in the multiple interested topics, so that the automatic determination of the target topics can be realized, manual screening is not needed, human resources are saved, and the topic recommendation efficiency is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of an information recommendation method according to an embodiment of the present invention;
fig. 2 is a flow chart of an information recommendation method according to an embodiment of the present invention;
fig. 3 is a schematic flow chart of an information recommendation method according to an embodiment of the present invention;
fig. 4 is a flow chart of an information recommendation method according to an embodiment of the present invention;
fig. 5 is a flow chart of an information recommendation method according to an embodiment of the present invention;
fig. 6 is a flow chart of an information recommendation method according to an embodiment of the present invention;
fig. 7 is a schematic structural diagram of an information recommendation device according to an embodiment of the present invention;
fig. 8 is a schematic structural diagram of a server according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention.
Thus, the following detailed description of the embodiments of the present application, as provided in the accompanying drawings, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, are intended to be within the scope of the present application.
In the description of the present application, it should be noted that, if the terms "upper", "lower", and the like indicate an azimuth or a positional relationship based on the azimuth or the positional relationship shown in the drawings, or an azimuth or the positional relationship that is commonly put when the product of the application is used, it is merely for convenience of description and simplification of the description, and does not indicate or imply that the apparatus or element to be referred to must have a specific azimuth, be configured and operated in a specific azimuth, and therefore should not be construed as limiting the present application.
Furthermore, the terms first, second and the like in the description and in the claims and in the above-described figures, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
It should be noted that, without conflict, features in embodiments of the present application may be combined with each other.
In the related technology, a topic recommendation mode is to generate a core concept of a topic based on operation experience and popular trend definition by an operator, then manually screen out a program composition topic conforming to the defined core concept based on the defined core concept, and deploy topic configuration to a terminal to realize topic recommendation. However, in the related art, the programs need to be manually screened out to form the theme, which wastes unnecessary human resources and reduces the efficiency of theme recommendation.
Aiming at the technical problems in the related art, the information recommendation method provided by the embodiment of the application determines the preference label of the target user, determines a plurality of interested topics of the target user from a plurality of topics corresponding to the plurality of topic labels according to the plurality of topic labels and the preference label in the topic library, then determines a target topic from the plurality of interested topics according to video content in the plurality of interested topics, namely, a topic to be recommended, and recommends video content in the target topic to terminal equipment corresponding to the target user; the method can automatically determine the target theme without manual screening, saves human resources and improves the theme recommendation efficiency.
The embodiment of the application provides an information recommendation method which can be applied to a server, and the information recommendation method provided by the embodiment of the application is explained below.
Fig. 1 is a flow chart of an information recommendation method provided in an embodiment of the present invention, as shown in fig. 1, the method may include:
s101, determining preference labels of the target user from a plurality of labels of the historical video according to the operation behaviors of the target user on the historical video.
The historical video is the video watched by the target user in the history. The operational behavior may include: search behavior, play behavior, click behavior.
In some embodiments, a preference tag of the target user for each tag dimension is determined from a plurality of tags of a plurality of tag dimensions of the historical video according to the operational behavior of the target user for the historical video, wherein each tag dimension of the historical video has a plurality of tags.
In practical applications, the plurality of tag dimensions may include: primary classification, secondary classification, lead actor, director, year, region, language. Wherein, the label under the first class classification may include: TV play and movie; tags under the secondary classification may include: comedy, love, drama.
S102, determining a plurality of interesting topics of the target user from a plurality of topics corresponding to the topic labels according to the topic labels and the preference labels in the topic library.
The topic library may include a plurality of topic tags, and a plurality of topics corresponding to each topic tag, where each topic has a corresponding topic id (Identity document, unique code) and a corresponding topic name.
In some embodiments, a matching result may be obtained for a plurality of topic tags and preference tags in the topic library, and a plurality of topics of interest of the target user may be determined from a plurality of topics corresponding to the topic tags according to the matching result.
It should be noted that, the matching may be performed on a plurality of theme labels and preference labels sequentially, or may be performed on a plurality of theme labels and preference labels simultaneously, or may be performed on a plurality of theme labels and preference labels in other manners, which is not limited in this embodiment of the present application.
S103, determining a target theme from the multiple interesting themes according to video content in the multiple interesting themes.
Wherein the number of target subjects may be at least one.
In some embodiments, a preset topic screening rule is adopted, and a plurality of topics of interest are screened according to video content in the plurality of topics of interest to obtain at least one target topic, wherein each target topic can comprise at least one video content.
In addition, the video content may include successive multi-frame images as well as audio.
S104, recommending the video content in the target theme to the terminal equipment corresponding to the target user.
The server can send the video content in the target theme to the terminal equipment corresponding to the target user, and the terminal equipment corresponding to the target user receives the video content in the target theme and then presents the video content in the target theme to the target user.
In this embodiment of the present application, the terminal device corresponding to the target user may be presented with a graphical user interface, where the graphical user interface includes a theme recommendation control, and the terminal device corresponding to the target user generates and sends a theme recommendation request to the server in response to a selection operation for the theme recommendation control input by the target user, and the server may execute the processes from S101 to S104 according to the theme recommendation request.
In addition, the terminal device corresponding to the target user responds to the sliding operation input by the target user, performs sliding display on the graphical user interface, and if the terminal device slides to the theme recommendation area in the graphical user interface, generates and sends a theme recommendation request to the server, and the server can execute the processes from S101 to S104 according to the theme recommendation request. Of course, the terminal device corresponding to the target user may also generate the theme recommendation request in other manners, which is not limited in particular in the embodiment of the present application.
In summary, an embodiment of the present invention provides an information recommendation method, including: determining a preference tag of the target user from a plurality of tags of the historical video according to the operation behavior of the target user on the historical video, wherein the historical video is the video watched by the target user in the history; determining a plurality of interesting topics of the target user from a plurality of topics corresponding to the topic labels according to the topic labels and the preference labels in the topic library; determining a target theme from the plurality of topics of interest according to video content in the plurality of topics of interest; and recommending the video content in the target theme to the terminal equipment corresponding to the target user. The preference labels of the target users and the multiple interested topics of the target users are automatically determined, and then the target topics are determined from the multiple interested topics according to video content in the multiple interested topics, so that the automatic determination of the target topics can be realized, manual screening is not needed, human resources are saved, and the topic recommendation efficiency is improved.
Optionally, fig. 2 is a flowchart of an information recommendation method provided in the embodiment of the present invention, as shown in fig. 2, a process of determining, in S101, a preference label of a target user from a plurality of labels of a history video according to an operation behavior of the target user with respect to the history video may include:
S201, determining preference scores of the target user for a plurality of labels of each label dimension according to the operation behaviors of the target user for the historical video.
Wherein the number of tag dimensions may be a plurality.
Optionally, the preference scores for the plurality of tags in the plurality of tag dimensions may include: the preference scores of the plurality of tags in the first class, the preference scores of the plurality of tags in the second class, the preference scores of the plurality of tags in the director, the preference scores of the plurality of tags in the year, the preference scores of the plurality of tags in the region, the preference scores of the plurality of tags in the language.
In addition, the server may sequentially calculate the preference scores of the target user for the plurality of tags, may calculate the preference scores of the target user for the plurality of tags at the same time, and may calculate the preference scores of the target user for the plurality of tags in other manners.
S202, sorting the labels of each label dimension according to preference scores of the labels of each label dimension, and obtaining sorted labels of each label dimension.
In some embodiments, the plurality of tags in each tag dimension may be sorted in descending order according to the preference scores of the plurality of tags, resulting in a sorted plurality of tags; the labels in each label dimension may also be sorted in ascending order according to the preference scores of the labels, resulting in a sorted plurality of labels.
S203, determining preference labels of each label dimension from the plurality of ordered labels of each label dimension.
In the embodiment of the application, a first number of tags with highest preference scores in the plurality of tags after the ordering of each tag dimension is used as preference tags of each tag dimension. Wherein the first number is denoted as k.
Optionally, fig. 3 is a flowchart of an information recommendation method provided in the embodiment of the present invention, as shown in fig. 3, a process of determining, according to an operation behavior of a target user on a historical video, preference scores of the target user on a plurality of labels of each label dimension in S201 may include:
s301, for each historical time period, obtaining an initial score of a target user for each tag in each historical time period according to the behavior type of the operation behavior, the behavior weight of the behavior type, the number of times the target user performs the operation behavior for each tag in a plurality of tags in each tag dimension, the number of times the target user performs the operation behavior for all tags in the target tag dimension corresponding to each tag, the number of times the target user performs the operation behavior for all tags in the target tag dimension, and the number of times the target user performs the operation behavior for each tag.
The behavior types may include search behavior type, play behavior type, click behavior type, among others. The plurality of historical time periods can be a plurality of historical time periods, each historical time period can be expressed as T, the operation behaviors of different period dimensions are counted, and the scheme is combined with long-term and short-term behavior data of a user to conduct recommendation in consideration of the comprehensiveness of theme recommendation. By way of example, the plurality of historical time periods may include: four historical time periods of last 1 day, last 7 days, last 15 days, last 30 days, etc.
S302, accumulating the initial scores of each historical time period to obtain preference scores of the target user for each tag.
Wherein, a preset first formula may be used to calculate a preference score of the target user for each tag.
In some embodiments, the first formula may be:
note that Score uj Representing the preference score of target user u for tag j, A represents the user's behavior type set, w a Behavior weight, cnt, representing behavior type a tuja The number of times of the occurrence behavior type a of the target user u to the tag j is represented by taking the behavior data with the history time period of t. all_cnt tua And (3) representing the times of taking the behavior data with the historical time period of t and taking the behavior type of a of all tags under the dimension of the target tag corresponding to the tag j by the target user u. all_cnt ta And (3) representing the times of taking the operation behaviors with the historical time period of t and taking the types of the behaviors of all the labels in the dimension of the target label corresponding to the label j as a number of times. cnt tja The operation behavior with the history time period being t is expressed, and the number of times that the occurrence behavior of the label j by the whole users is a is expressed.
Notably, w a Representing the behavior weight of behavior type a. Different behavior types are different in reflecting the preference degree of the user, such as searching and playing behaviors are more important than clicking, so that higher weight is also supposed to be obtained, and the configuration is needed according to actual service conditions and service experience.
In addition, relationship of behavior weight to final recommendation effect: since the behavior weights will participate in the modeling score calculation of the user's interests, if the user experiences a higher-weighted behavior for video a, the algorithm will assign higher recommendation weights to other videos a ' that are similar to a, and thus will prefer to recommend video a '. Therefore, the target behaviors (such as playing > searching > clicking) expected to occur by the user in the service can be set with higher weight in a targeted manner, so that a better recommendation effect is obtained.
Optionally, fig. 4 is a flowchart of an information recommendation method provided in the embodiment of the present invention, as shown in fig. 4, a process of determining, in S102, a plurality of topics of interest of a target user from a plurality of topics corresponding to a plurality of topic labels according to a plurality of topic labels and preference labels in a topic library may include:
S401, matching the theme labels and the preference labels to obtain interest scores of the target user for the theme.
In some embodiments, a preset second formula may be adopted to match the plurality of topic tags and preference tags, so as to obtain interest scores of the target user for the plurality of topics.
S402, sorting the topics according to interest scores of the topics to obtain sorted topics.
In the embodiment of the application, the topics are sorted in descending order according to the interest scores of the topics to obtain sorted topics, or sorted in ascending order according to the interest scores of the topics to obtain sorted topics.
S403, determining a plurality of interesting topics of the target user from the sorted topics.
Wherein, a plurality of interesting topics form a topic list, namely, topic recalls are carried out.
It should be noted that, from the multiple topics after sorting, the second number of topics with the highest interest score is selected as multiple topics of interest of the target user. Wherein the second number may be identified as N. By way of example, N may be 100.
Optionally, the process of matching the plurality of topic tags and the preference tag to obtain the interest scores of the target user for the plurality of topics in S401 may include:
And calculating according to the preset weight corresponding to the label dimension, the matching quantity of the preference label and each theme label in the corresponding label dimension and the corresponding preset value in the label dimension to obtain interest scores of a plurality of themes.
In some embodiments, the preset second formula may be:
wherein Score us Representing the interest score of the target user u for the topic s. L represents a set of tag dimensions, which may include primary classification, secondary classification, lead, director, year, region, language, etc. w (w) l The preset weight corresponding to the label dimension l is represented, and can be also called label matching weight; and configuring different tag matching weights for the tags in different tag dimensions according to actual service conditions and service experience. cnt l The preference label representing the target user u is paired with the topic label of the topic sNumber of matches in the label dimension l. k (k) l The preset value representing the label dimension l may be a first number, i.e. the number of preferred labels therein when the label dimension is l.
For example, the tag matching weight for the primary class is 100 and the tag matching weight for the secondary class is 80. The primary class label k is 2 and the secondary class k is 3. The first class classification preference of the target user u is [ TV drama, movie ], and the second class classification preference is [ love, action, comedy ]. The subject s tag is: the primary class label is a movie and the secondary class label is [ comedy |love ]. The user's preference label matches the subject information label with a score of 100 x 1/2+80 x 2/3= 103.33.
Optionally, fig. 5 is a flowchart of an information recommendation method provided in an embodiment of the present invention, as shown in fig. 5, a process of determining, in S103, a target topic from a plurality of topics of interest according to video content in the plurality of topics of interest may include:
s501, scoring the video content in each interested subject to obtain the score of each video content in each interested subject.
In the embodiment of the application, a preset scoring rule is adopted, and each interested subject has a plurality of video contents, so that each video content in each interested subject can be scored to obtain the score of each video content in each interested subject.
S502, determining target video content of each interested subject according to the score of each video content in each interested subject.
Wherein the target video content for each topic of interest may be referred to as a pool of program content for each topic of interest.
In some embodiments, the plurality of video contents in each topic of interest are ordered in descending order according to the score of each video content in each topic of interest, and a third number of video contents with the highest score in each topic of interest is selected as the target video content of each topic of interest, where the third number may be denoted as m.
S503, obtaining the score of each interested topic according to the score of the target video content of each interested topic.
S504, determining a target theme from a plurality of interesting themes according to the score of each interesting theme.
In some implementations, an average of the scores of the plurality of target video content for each topic of interest is calculated, resulting in a score for each topic of interest; and sorting the multiple topics of interest in a descending order according to the score of each interest, and taking the fourth number of topics of interest with the highest score in the multiple topics of interest as target topics. Wherein the third number may be denoted as n.
Illustratively, n may be 100 and m may be 10.
Optionally, fig. 6 is a flowchart of an information recommendation method provided in the embodiment of the present invention, as shown in fig. 6, a process of scoring video content in each topic of interest in S501 to obtain a score of each video content in each topic of interest may include:
s601, obtaining characteristic information of a target user and characteristic information of video content in each interested subject.
And extracting the characteristics of the target user and the video content in each interested subject to obtain the characteristic information of the target user and the characteristic information of the video content in each interested subject.
In the embodiment of the present application, the feature information of the target user may include: basic attributes of the target user (age, gender, region, device information, whether members are or not, etc.), behavior sequences (viewing sequences), interest preferences, i.e., preference tags in various tag dimensions (primary classification preference, secondary classification preference, director preference, language preference, region preference, etc. for videos), liveness and viscosity (last access time, number of views, frequency).
In addition, the feature information of the video content in each of the topics of interest may include: basic attributes and a welcome program, wherein the basic attributes may include: video id, primary and secondary classification, lead, director, language, popular program may include: number of plays, uv (texture map coordinates), click rate, play conversion rate, time average play duration, person average play duration.
S602, processing the characteristic information of the target user and the characteristic information of the video content in each interested subject by adopting a preset video content scoring model to obtain the score of each video content in each interested subject.
The score of each video content in each interested subject can be used for representing the preset on-demand rate of each video content in each interested subject.
In some embodiments, a preset video content scoring model is adopted, feature information of a target user and feature information of video content in each interested subject are extracted as vectors, the vectors are spliced, the spliced vectors are input into a multi-layer DNN (Deep Neural Network ), and the DNN can output the score of each video content in each interested subject. The DNN is a network structure in a preset video content scoring model.
Optionally, the method may further include:
and performing model training according to the characteristic information of the sample user, the first sample video content which is exposed and requested by the sample user, and the second sample video content which is exposed and not requested by the sample user, so as to obtain a preset video content scoring model.
Wherein the first sample video content is a positive sample and the second sample video content is a negative sample.
The first sample video content and the second sample video content are processed to obtain the characteristic information of the first sample video content and the characteristic information of the second sample video content. Model training is carried out according to the characteristic information of the sample user, the characteristic information of the first sample video content and the characteristic information of the second sample video content, and a preset video content scoring model is obtained.
In addition, the preset video content scoring model may include: input layer, embedding layer, mlp (perceptron) layer, output layer connected in turn.
In the embodiment of the application, the characteristic information of the sample user is as follows: basic attributes of sample users (age, gender, region, device information, whether members are etc.), behavior sequences (viewing sequences), interest preferences, i.e., preference tags in various tag dimensions (primary classification preference, secondary classification preference, director preference, language preference, region preference, etc. for video), liveness and viscosity (last visit from present duration, number of views, frequency).
The feature information of the first sample video content may include: basic attributes and a welcome program, wherein the basic attributes may include: video id, primary and secondary classification, lead, director, language, popular program may include: number of plays, uv (texture map coordinates), click rate, play conversion rate, time average play duration, person average play duration.
It should be noted that, if the target user is a new user without searching, playing and clicking actions, the topical theme and topical video content in the program content pool of the theme are taken as default recommended theme and theme content and presented to the terminal user. If the target user is an old user with searching, playing and clicking actions, the personalized recommendation result is presented to the terminal user based on the video content of the target subject in the step S104. And for the situation that the recommended number does not meet the recommended number requirement, the obtained default recommended topics and the complemented topic contents are presented to the terminal equipment corresponding to the target user.
In summary, by collecting behavior data of the target user, interest figures of the target user on video content, labels of the video content and topics are depicted. A fixed number of topic lists are recalled by matching topic tags with preference tags of the target user. And screening a program content pool of topics through topic labels, predicting interest scores of target users on video content based on a deep learning model, obtaining a fixed number of topic contents which are more interested by the target users for each topic, and finally obtaining a topic list which is more interested by the target users by aggregating the interest scores of the target users on the topic contents. Therefore, interested topics and topic contents are recommended to target users in real time, personalized topic recommendation effects of thousands of people and thousands of faces are achieved, target user experience is improved, recommendation efficiency is improved, and operation burden is reduced.
The following describes an information recommendation device, a server, a storage medium, etc. for executing the information recommendation method provided in the present application, and specific implementation processes and technical effects thereof refer to relevant contents of the information recommendation method, which are not described in detail below.
Fig. 7 is a schematic structural diagram of an information recommendation device according to an embodiment of the present invention, where, as shown in fig. 7, the device may include:
A determining module 701, configured to determine, according to an operation behavior of a target user on a historical video, a preference tag of the target user from a plurality of tags of the historical video, where the historical video is a video that the target user historically watched; determining a plurality of interesting topics of the target user from a plurality of topics corresponding to the topic tags according to the topic tags and the preference tags in the topic library; determining a target theme from the plurality of topics of interest according to video content in the plurality of topics of interest;
and the recommending module 702 is used for recommending the video content in the target theme to the terminal equipment corresponding to the target user.
Optionally, the determining module 701 is specifically configured to determine, according to the operation behavior of the target user for the historical video, preference scores of a plurality of tags of the target user for each tag dimension; sorting the labels of each label dimension according to the preference scores of the labels of each label dimension to obtain sorted labels of each label dimension; and determining the preference label of each label dimension from the plurality of ordered labels of each label dimension.
Optionally, the determining module 701 is specifically configured to obtain, for each historical time period, an initial score of the target user for each tag in the historical time period according to a behavior type of the operation behavior, a behavior weight of the behavior type, a number of times the target user performs the operation behavior for each tag in a plurality of tags in each tag dimension, a number of times the target user performs the operation behavior for all tags in a target tag dimension corresponding to the each tag, a number of times the target user performs the operation behavior for all tags in the target tag dimension, and a number of times the target user performs the operation behavior for each tag; and accumulating the initial scores of each historical time period to obtain preference scores of the target user for a plurality of labels of each label dimension.
Optionally, the determining module 701 is specifically configured to match the plurality of topic tags and the preference tag to obtain interest scores of the target user for the plurality of topics; sorting the topics according to the interest scores of the topics to obtain sorted topics; and determining a plurality of interesting topics of the target user from the ordered topics.
Optionally, the determining module 701 is specifically configured to calculate, according to a preset weight corresponding to a tag dimension, a number of matches between the preference tag and each topic tag in the corresponding tag dimension, and a preset value corresponding to the tag dimension, to obtain interest scores of the multiple topics.
Optionally, the determining module 701 is specifically configured to score video content in each topic of interest, so as to obtain a score of each video content in each topic of interest; determining target video content of each interested subject according to the score of each video content in each interested subject; obtaining the score of each interested subject according to the score of the target video content of each interested subject; and determining a target theme from the multiple themes of interest according to the score of each theme of interest.
Optionally, the determining module 701 is specifically configured to obtain feature information of the target user and feature information of video content in each topic of interest; and processing the characteristic information of the target user and the characteristic information of the video content in each interested subject by adopting a preset video content scoring model to obtain the score of each video content in each interested subject.
Optionally, the apparatus further includes:
and the training module is used for carrying out model training according to the characteristic information of the sample user, the first sample video content which is exposed and requested by the sample user, and the second sample video content which is exposed and not requested by the sample user, so as to obtain the preset video content scoring model.
The foregoing apparatus is used for executing the method provided in the foregoing embodiment, and its implementation principle and technical effects are similar, and are not described herein again.
The above modules may be one or more integrated circuits configured to implement the above methods, for example: one or more application specific integrated circuits (Application Specific Integrated Circuit, abbreviated as ASIC), or one or more microprocessors (digital singnal processor, abbreviated as DSP), or one or more field programmable gate arrays (Field Programmable Gate Array, abbreviated as FPGA), or the like. For another example, when a module above is implemented in the form of a processing element scheduler code, the processing element may be a general-purpose processor, such as a central processing unit (Central Processing Unit, CPU) or other processor that may invoke the program code. For another example, the modules may be integrated together and implemented in the form of a system-on-a-chip (SOC).
Fig. 8 is a schematic structural diagram of a server according to an embodiment of the present invention, as shown in fig. 8, where the server may include: a processor 801, and a memory 802.
The memory 802 is used for storing a program, and the processor 801 calls the program stored in the memory 802 to execute the above-described method embodiment. The specific implementation manner and the technical effect are similar, and are not repeated here.
Optionally, the present invention also provides a program product, such as a computer readable storage medium, comprising a program for performing the above-described method embodiments when being executed by a processor.
In the several embodiments provided by the present invention, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in hardware plus software functional units.
The integrated units implemented in the form of software functional units described above may be stored in a computer readable storage medium. The software functional unit is stored in a storage medium, and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor (english: processor) to perform some of the steps of the methods according to the embodiments of the invention. And the aforementioned storage medium includes: u disk, mobile hard disk, read-Only Memory (ROM), random access Memory (Random Access Memory, RAM), magnetic disk or optical disk, etc.
The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. An information recommendation method, comprising:
determining a preference tag of a target user from a plurality of tags of a historical video according to the operation behavior of the target user on the historical video, wherein the historical video is a video which is watched by the target user in a historical manner;
determining a plurality of interesting topics of the target user from a plurality of topics corresponding to the topic tags according to the topic tags and the preference tags in the topic library;
determining a target theme from the plurality of topics of interest according to video content in the plurality of topics of interest;
and recommending the video content in the target theme to the terminal equipment corresponding to the target user.
2. The method of claim 1, wherein the determining the preference tag of the target user from the plurality of tags of the historical video according to the operational behavior of the target user with respect to the historical video comprises:
Determining preference scores of the target user for a plurality of labels of each label dimension according to the operation behaviors of the target user for the historical video;
sorting the labels of each label dimension according to the preference scores of the labels of each label dimension to obtain sorted labels of each label dimension;
and determining the preference label of each label dimension from the plurality of ordered labels of each label dimension.
3. The method of claim 2, wherein the determining the preference scores of the target user for the plurality of tags for each tag dimension based on the operational behavior of the target user for the historical video comprises:
for each historical time period, obtaining an initial score of the target user for each tag in each historical time period according to the behavior type of the operation behavior, the behavior weight of the behavior type, the times of the operation behavior of the target user for each tag in a plurality of tags of each tag dimension, the times of the operation behavior of the target user for all tags in the target tag dimension corresponding to each tag, the times of the operation behavior of the target user for all tags in the target tag dimension, and the times of the operation behavior of the target user for each tag;
And accumulating the initial scores of each historical time period to obtain preference scores of the target user for a plurality of labels of each label dimension.
4. The method according to claim 1, wherein the determining, from a plurality of topics corresponding to the plurality of topic tags, a plurality of topics of interest of the target user according to the plurality of topic tags in the topic library and the preference tag comprises:
matching the theme labels and the preference labels to obtain interest scores of the target user for the theme labels;
sorting the topics according to the interest scores of the topics to obtain sorted topics;
and determining a plurality of interesting topics of the target user from the ordered topics.
5. The method of claim 4, wherein the matching the plurality of topic tags with the preference tag to obtain interest scores of the target user for the plurality of topics comprises:
and calculating according to the preset weight corresponding to the label dimension, the matching quantity of the preference label and each theme label in the corresponding label dimension and the corresponding preset value in the label dimension to obtain interest scores of the themes.
6. The method of claim 1, wherein the determining a target topic from the plurality of topics of interest based on video content in the plurality of topics of interest comprises:
scoring the video content in each interested subject to obtain the score of each video content in each interested subject;
determining target video content of each interested subject according to the score of each video content in each interested subject;
obtaining the score of each interested subject according to the score of the target video content of each interested subject;
and determining a target theme from the multiple themes of interest according to the score of each theme of interest.
7. The method of claim 6, wherein scoring the video content in each topic of interest to obtain a score for each video content in each topic of interest comprises:
acquiring the characteristic information of the target user and the characteristic information of the video content in each interested subject;
and processing the characteristic information of the target user and the characteristic information of the video content in each interested subject by adopting a preset video content scoring model to obtain the score of each video content in each interested subject.
8. The method of claim 7, wherein the method further comprises:
and performing model training according to the characteristic information of the sample user, the first sample video content which is exposed and requested by the sample user, and the second sample video content which is exposed and not requested by the sample user, so as to obtain the preset video content scoring model.
9. An information recommendation device, characterized by comprising:
the determining module is used for determining preference tags of the target user from a plurality of tags of the historical video according to the operation behaviors of the target user on the historical video, wherein the historical video is the video which is historically watched by the target user; determining a plurality of interesting topics of the target user from a plurality of topics corresponding to the topic tags according to the topic tags and the preference tags in the topic library; determining a target theme from the plurality of topics of interest according to video content in the plurality of topics of interest;
and the recommending module is used for recommending the video content in the target theme to the terminal equipment corresponding to the target user.
10. A server, comprising: a memory and a processor, the memory storing a computer program executable by the processor, the processor implementing the information recommendation method of any of the preceding claims 1-8 when the computer program is executed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310134565.8A CN116112710A (en) | 2023-02-17 | 2023-02-17 | Information recommendation method, device and server |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310134565.8A CN116112710A (en) | 2023-02-17 | 2023-02-17 | Information recommendation method, device and server |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116112710A true CN116112710A (en) | 2023-05-12 |
Family
ID=86254037
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310134565.8A Pending CN116112710A (en) | 2023-02-17 | 2023-02-17 | Information recommendation method, device and server |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116112710A (en) |
-
2023
- 2023-02-17 CN CN202310134565.8A patent/CN116112710A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106326391B (en) | Multimedia resource recommendation method and device | |
| CN111708901B (en) | Multimedia resource recommendation method and device, electronic equipment and storage medium | |
| CN111444428A (en) | Information recommendation method and device based on artificial intelligence, electronic equipment and storage medium | |
| CN108595493B (en) | Media content pushing method and device, storage medium and electronic device | |
| CN111611436A (en) | Label data processing method and device and computer readable storage medium | |
| CN112597395A (en) | Object recommendation method, device, equipment and storage medium | |
| CN112507163B (en) | Duration prediction model training method, recommendation method, device, equipment and medium | |
| US20210044870A1 (en) | Representation Of Content Based On Content-Level Features | |
| CN114095749B (en) | Recommendation and live interface display method, computer storage medium and program product | |
| CN102884524A (en) | Automatic image discovery and recommendation for displayed television content | |
| CN113742567B (en) | Recommendation method and device for multimedia resources, electronic equipment and storage medium | |
| CN104685899A (en) | Dynamic media segment pricing | |
| CN111723237A (en) | Media content access control method | |
| US11172271B1 (en) | Automated generation of banner images | |
| CN111711869A (en) | Label data processing method and device and computer readable storage medium | |
| CN111597446A (en) | Content pushing method and device based on artificial intelligence, server and storage medium | |
| CN113420166A (en) | Commodity mounting, retrieving, recommending and training processing method and device and electronic equipment | |
| CN114845149B (en) | Video clip method, video recommendation method, device, equipment and medium | |
| CN106445922B (en) | Method and device for determining title of multimedia resource | |
| CN116112710A (en) | Information recommendation method, device and server | |
| CN110381339B (en) | Picture transmission method and device | |
| CN112533032B (en) | Video data processing method and device and storage medium | |
| CN114357301A (en) | Data processing method, device and readable storage medium | |
| KR101452414B1 (en) | Method for providing multimedia contents using meta information | |
| CN114969493A (en) | Content recommendation method and related device |
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 |