CN110430471B - Television recommendation method and system based on instantaneous calculation - Google Patents
Television recommendation method and system based on instantaneous calculation Download PDFInfo
- Publication number
- CN110430471B CN110430471B CN201910672136.XA CN201910672136A CN110430471B CN 110430471 B CN110430471 B CN 110430471B CN 201910672136 A CN201910672136 A CN 201910672136A CN 110430471 B CN110430471 B CN 110430471B
- Authority
- CN
- China
- Prior art keywords
- user
- recommendation
- matrix
- data
- content
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
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]
- H04N21/40—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
- H04N21/4661—Deriving a combined profile for a plurality of end-users of the same client, e.g. for family members within a home
-
- 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
Abstract
The invention discloses a television recommendation method and system based on instantaneous calculation, wherein the method comprises the following steps: acquiring article content data and user behavior data, and constructing a content matrix and a user matrix; performing hierarchical text classification on the content matrix, and establishing a knowledge graph; performing family portrait modeling on a user matrix; establishing a recommendation model based on the content matrix and the user matrix; according to the recommendation model, performing initial program recommendation to a user based on the current article content data; and receiving behavior data of the user aiming at the recommendation result, performing instantaneous calculation on the video recommendation model based on reinforcement learning, correcting the recommendation model, and updating the recommendation result. The method is applied to IPTV television recommendation, and can self-adaptively obtain and meet the requirements of users watching the television at present without user distinction.
Description
Technical Field
The invention belongs to the technical field of intelligent content recommendation, and particularly relates to a television recommendation method and system based on instantaneous calculation.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
IPTV, also known as internet protocol television, is a video service built on an IP-based private broadband network that can provide high quality traditional television channel programming and audio-video on-demand content. Compared with the traditional television program, the IPTV has the advantages that not only can real-time on-line programs same as the traditional television be watched, but also programs which the IPTV wants to watch can be searched for on-line watching through a data platform provided by a service provider. However, the appearance of a large number of tv programs also makes it difficult for users to accurately and quickly index their favorite tv programs among a huge number of tv programs, so it is necessary to intelligently recommend the tv programs that the users are interested in.
However, although IPTV is also a video recommendation, it is substantially different from the conventional video recommendation. Traditional network video recommendation such as websites of Youku, potato, love art and the like obtains user preference by collecting information of watching logs, collections, comments and the like of individuals, and is easy to realize. However, in IPTV video recommendation, personal preference information of a user cannot be collected based on behaviors such as collection and comments, and only viewing logs of the user can be relied on, and IPTV is often a group of family users for which viewing records of different family members are mixed together, and thus targeted recommendation is difficult. Relevant researchers such as the royal talent distinguish family members by dividing time periods, but the inventor finds that the pertinence and the accuracy of recommendation in the way are to be improved.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a television recommendation method and system based on instantaneous calculation.
In order to achieve the above object, one or more embodiments of the present invention provide the following technical solutions:
a method for tv recommendation based on instantaneous calculation, comprising the steps of:
acquiring article content data and user behavior data, and constructing a content matrix and a user matrix;
performing hierarchical text classification on the content matrix, and establishing a knowledge graph;
performing family portrait modeling on a user matrix;
establishing a recommendation model based on the content matrix and the user matrix;
according to the recommendation model, performing initial program recommendation to a user based on the current article content data;
and receiving behavior data of the user aiming at the recommendation result, recalling the recommendation result based on the reinforcement learning instantaneous calculation engine, correcting the recommendation model through calculating loss, and updating the recommendation result.
One or more embodiments provide an instantaneous computation-based television recommendation system, comprising:
the data acquisition module is used for acquiring article content data and user behavior data and constructing a content matrix and a user matrix;
the data processing module is used for carrying out hierarchical text classification on the content matrix and establishing a knowledge map; performing family portrait modeling on a user matrix;
the recommendation model establishing module is used for establishing a recommendation model based on the content matrix and the user matrix;
the initial program recommending module is used for recommending the initial program to the user based on the current article content data according to the recommending model;
and the instantaneous calculation module is used for receiving behavior data of a user aiming at the recommendation result, recalling the recommendation result based on the instantaneous calculation engine of the reinforcement learning, correcting the recommendation model through calculating loss and updating the recommendation result.
One or more embodiments provide an electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the instant computing-based television recommendation method when executing the program.
One or more embodiments provide a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the instant computing-based television recommendation method.
The above one or more technical solutions have the following beneficial effects:
different from the traditional IPTV video recommendation system, the recommendation system does not distinguish the family users, and after the behavior data of the current user is obtained, the recommendation model is quickly corrected and finely adjusted through instantaneous calculation, so that the recommendation result is obtained.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
FIG. 1 is a flow diagram illustrating a method for transient-computing-based television recommendation in accordance with one or more embodiments of the present invention; fig. 2 is a schematic flow chart illustrating a tv recommendation method based on instantaneous computation according to one or more embodiments of the present invention.
Detailed Description
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
The embodiments and features of the embodiments of the present invention may be combined with each other without conflict.
Example one
The embodiment discloses a television recommendation method based on instantaneous calculation, which comprises two stages of off-line and on-line, as shown in FIGS. 1-2.
The off-line stage comprises steps 1-3, specifically:
step 1: acquiring article content data and user data;
the item data includes: content provided by a content provider (e.g., EPG), and a program guide, etc.;
the user data includes: the IPTV system comprises an IPTV user number, behavior starting time, behavior types, program watching duration and program information, wherein the behavior type data comprises browsing, on-demand, live broadcast and collection, and the program types comprise a primary type to which a program belongs and a secondary type to which the program belongs; the program information includes program number, program name, program type, duration of the program itself, director, and the like.
Step 2: preprocessing the content data and the user data of the article;
the method specifically comprises the steps of data dimension reduction, data cleaning and construction of a knowledge graph and a family portrait. Specifically, text processing is carried out on the content data of the article, stop words are taken out, hidden semantic analysis is carried out to obtain text characteristics, data integration and knowledge extraction are carried out through hierarchical text classification, and a knowledge graph is established; and for the user behavior data, obtaining the scores of all the contents, and constructing a score matrix based on the scores and the corresponding programs. Wherein the scoring comprises explicit scoring and implicit scoring.
In this embodiment, implicit rating is adopted, for IPTV live programs, rating is proportional to user viewing time, and programs with playing time less than a set time (5 minutes) are rejected, and for other programs, a rating r calculation method is as follows:
wherein t represents the watching time length, and L is the total time length of the program.
For video programs of the on-demand type, the system automatically assigns a higher score (e.g., 5 points full and 4 points) as an implicit score whenever the user requests to watch a movie.
One half of full score is subtracted from all scores (e.g., 5 points full, 2.5 points subtracted) to distinguish between positive and negative scores.
And respectively storing the content characteristics and the user scores into a content matrix (ICM) and a user score matrix (URM) for establishing a subsequent recommendation model.
The article content data comprises structure data and semi-structure data, and for the structure data, data fusion is required to be carried out with the program data; for semi-structural data, operations such as entity extraction, attribute extraction, relation extraction and the like are required to be carried out, coreference resolution is carried out together with a result obtained after data fusion, then hierarchical text classification and index creation are carried out, and classified texts are stored in a content master database according to indexes.
The user behavior data comprises business data, log data, buried point data, external data and the like, for the received user behavior data, ETL processing is carried out on a data warehouse, classification of the user behavior data is executed based on machine learning algorithms such as logistic regression, random decision trees, LDA models and the like, finally a user image is created based on user attributes, behaviors, preferences, numbers and the like, and the data is stored in a user master database.
Thus, the IPTV television watching habit of a certain family, namely a family picture, is obtained, which is used for describing the preference of the family, wherein the family picture comprises the television categories (a primary category and a secondary category can be subdivided) which are preferred to watch, and the scores of the categories. The television category reflects the preference of each member of the family, and can imply which members the family contains, for example, animation films are contained in the television category, which means that the family has children; the grade of each category can reflect the playing amount of each category, and the time for each family member to watch television is implied, so that the weight of each television category can be understood.
And step 3: carrying out batch processing on the article content data and the user behavior data to obtain a recommendation model, and storing the recommendation model in a model warehouse;
the calculation method of the storage decision system in the model warehouse comprises a classification algorithm, a clustering algorithm, an association rule algorithm, a collaborative filtering method, content-based recommendation and the like.
Currently, common recommendation methods include content-based recommendations, user-based collaborative filtering recommendations, program-based collaborative filtering recommendations, and model-based collaborative filtering recommendations. Wherein, based on the Collaborative Filtering Recommendation (User-based Collaborative Filtering): the essence of the collaborative filtering recommendation algorithm based on the users is to find users with the same favorite orientation as the target user, and after finding out similar users, recommend the target user according to the favorite degree of the similar users to the articles. Program-based Collaborative Filtering Recommendation (Item-based Collaborative Recommendation): according to the evaluation of the user on the articles or the information, the similarity between the articles is found, and then the similar articles are recommended to the user according to the historical preference of the user. In this embodiment, a recommendation model is established by fusing a recommendation based on content and a collaborative filtering recommendation based on a user, and specifically, the recommendation model establishment process includes:
(1) calculating the similarity between the target user and the nearest neighbor according to the initial user scoring matrix to obtain the neighbor with similar watched programs;
wherein, Iu,vRepresenting a set of programs rated by user u and neighbors v simultaneously, IuAnd IvRespectively representing the program sets scored by the user u and the neighbor v; r isu,iAnd rv,iRespectively representing the scores of user u and neighbor v for program i,
and
representing the average score of user u and neighbor v.
(2) Calculating the similarity between the target user preference and the program content characteristics watched by a plurality of nearest neighbors;
the similarity between the content features is calculated by means of the similarity of the text, wherein,
a content feature vector representing nearest neighbor broadcast program, U { (U)1,w1),(u2,w2),...,(un,wn) The user interest content feature vector is represented.
(3) Carrying out weighted fusion on the similarity obtained in the first two steps to obtain the total similarity sim and the similarity with a plurality of most similar neighbors;
(4) predicting program scores based on the total similarity sim, and performing descending order on the program scores to generate recommendation results:
firstly, selecting k users most similar to a target user according to the total similarity sim; then, the prediction score of the user u for the item i is calculated:
where V is a set of k similar users, rv,iIs the rating of the item i by the user v,
and
represents the average score of u and v. This formula first computes all neighbor preferences weighted by similarity and then adds the average score of the target user. And once the prediction score is obtained, the item with the highest score can be selected and recommended to the user.
The online phase comprises:
and 4, step 4: calling a recommendation model from a model warehouse based on the current article content data, and recommending an initial program to a user;
and feeding back the recommended program list to the user in modes of IPTV, mobile TV, Internet TV, limited TV and the like.
And 5: receiving behavior data of a user aiming at a recommendation result, performing instantaneous calculation on a video recommendation model based on reinforcement learning, correcting the recommendation model, and updating the recommendation result;
specifically, the online server records real-time behavior characteristics, imports a Kafka file queue and a Storm cluster, splices user complete data, constructs a sample, executes user behavior analysis, performs characteristic extraction on user behavior, and inputs watching records of the user into an instantaneous calculation engine. Each feedback action of the user is collected to serve as behavior data aiming at the recommendation result, after each behavior data is obtained, instantaneous calculation is carried out on the video recommendation model based on reinforcement learning, and corresponding reward is given to the neural network model based on reinforcement learning for positive feedback of the user so as to realize correction of the model and update the recommendation result.
If the moment when the user behavior is received is recorded as absolute time, a certain time difference exists between the updated result obtained through instantaneous calculation and the absolute time, and the time difference is recorded as relative time. The purpose of this embodiment is to update the model quickly by reinforcement learning, and shorten the relative time.
The behavior data includes all clicking behaviors performed by the user through a control device such as a remote controller, for example, operations such as selecting and watching a program, quitting and watching the program, and the behavior data is converted into behavior data of the user, for example: if the program in the initial program recommendation result is selected and watched, recording the related information and watching duration of the program; if the initial program recommendation result is rejected, searching other programs, recording related information of the program, and if the program is watched, recording watching time length.
Because the IPTV is oriented to a family user, when someone watches television each time, the user in front of the current television cannot be determined, by adopting the recommendation system of the embodiment, an initial recommendation list is recommended to the user, and by receiving the feedback behavior of the user, the model is rapidly adjusted based on instantaneous calculation, and the recommendation list is updated, so that the viewing interest of the current user can be rapidly adapted, and the user requirements are met. Compared with the traditional method of distinguishing users based on time periods and the like, the method adopted by the embodiment is simple to implement and more flexible, and can adaptively obtain and meet the requirements of users watching television at present without distinguishing the users.
Moreover, because reinforcement learning is essentially continuously corrected according to real-time operation of a user, news recommendation in the internet field and the like have poor reinforcement learning effects due to too sparse user behavior feedback, but the recommendation bit of the IPTV television is not large, so that the user behavior feedback is not sparse, and a better result can be obtained.
Example two
It is an object of the present embodiment to provide a television recommendation system based on instantaneous calculation.
To achieve the above object, the present embodiment provides a television recommendation system based on instantaneous calculation, including:
the data acquisition module is used for acquiring article content data and user behavior data and constructing a content matrix and a user matrix;
the data processing module is used for carrying out hierarchical text classification on the content matrix and establishing a knowledge map; performing family portrait modeling on a user matrix;
the recommendation model establishing module is used for establishing a recommendation model based on the content matrix and the user matrix;
the initial program recommending module is used for recommending the initial program to the user based on the current article content data according to the recommending model;
and the instantaneous calculation module is used for receiving behavior data of a user aiming at the recommendation result, recalling the recommendation result based on the instantaneous calculation engine of the reinforcement learning, correcting the recommendation model through calculating loss and updating the recommendation result.
EXAMPLE III
The embodiment aims at providing an electronic device.
An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the following steps when executing the program, comprising:
acquiring article content data and user behavior data, and constructing a content matrix and a user scoring matrix;
establishing a recommendation model based on the content matrix and the user scoring matrix;
according to the recommendation model, performing initial program recommendation to a user based on the current article content data;
and receiving behavior data of the user aiming at the recommendation result, performing instantaneous calculation on the video recommendation model based on reinforcement learning, correcting the recommendation model, and updating the recommendation result.
Example four
An object of the present embodiment is to provide a computer-readable storage medium.
A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:
acquiring article content data and user behavior data, and constructing a content matrix and a user scoring matrix;
establishing a recommendation model based on the content matrix and the user scoring matrix;
according to the recommendation model, performing initial program recommendation to a user based on the current article content data;
and receiving behavior data of the user aiming at the recommendation result, performing instantaneous calculation on the video recommendation model based on reinforcement learning, correcting the recommendation model, and updating the recommendation result.
The steps involved in the above second, third and fourth embodiments correspond to the first embodiment of the method, and the detailed description thereof can be found in the relevant description of the first embodiment. The term "computer-readable storage medium" should be taken to include a single medium or multiple media containing one or more sets of instructions; it should also be understood to include any medium that is capable of storing, encoding or carrying a set of instructions for execution by a processor and that cause the processor to perform any of the methods of the present invention.
One or more of the above embodiments have the following technical effects:
different from the traditional IPTV video recommendation system, the recommendation system does not distinguish the family users, and after the behavior data of the current user is obtained, the recommendation model is quickly corrected and finely adjusted through instantaneous calculation, so that the recommendation result is obtained.
Those skilled in the art will appreciate that the modules or steps of the present invention described above can be implemented using general purpose computer means, or alternatively, they can be implemented using program code that is executable by computing means, such that they are stored in memory means for execution by the computing means, or they are separately fabricated into individual integrated circuit modules, or multiple modules or steps of them are fabricated into a single integrated circuit module. The present invention is not limited to any specific combination of hardware and software.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Although the embodiments of the present invention have been described with reference to the accompanying drawings, it is not intended to limit the scope of the present invention, and it should be understood by those skilled in the art that various modifications and variations can be made without inventive efforts by those skilled in the art based on the technical solution of the present invention.
Claims (10)
1. A method for television recommendation based on transient calculations, comprising the steps of:
acquiring article content data and user behavior data, and constructing a content matrix and a user matrix; the user behavior data comprises service data, log data, buried point data and external data;
performing hierarchical text classification on the content matrix, and establishing a knowledge graph;
performing family portrait modeling on a user matrix;
establishing a recommendation model based on the content matrix and the user matrix;
according to the recommendation model, performing initial program recommendation to a user based on the current article content data;
behavior data of a user aiming at the recommendation result is received, the recommendation result is recalled based on an instantaneous calculation engine of reinforcement learning, the recommendation model is corrected through calculation loss, and the recommendation result is updated; the behavior data includes all clicking behaviors performed by the user through a control device such as a remote controller.
2. The method of claim 1 for transient-computing-based television recommendation, wherein the user scoring matrix is constructed by:
setting full scores of the scores, and calculating implicit scores of the programs by the user according to a rule that the scores are in direct proportion to the watching time of the user for IPTV live programs;
for on-demand video programs, whenever a user requests to watch a movie, the system automatically assigns a score as an implicit score;
and establishing a two-dimensional user scoring matrix based on the scores.
3. The method of claim 1 for transient computing based television recommendation, wherein the recommendation model building method comprises:
calculating the similarity between a target user and neighbors aiming at a user scoring matrix to obtain a plurality of neighbors which are most similar;
calculating the similarity of the target user preference and the content characteristics among the most similar neighbors;
carrying out weighted fusion on the two similarities obtained in the first two steps to obtain a total similarity;
program scores are predicted based on the total similarity and sorted in descending order to produce recommendations.
4. The method of claim 3, wherein the similarity between the target user and the neighbors is calculated by:
wherein, Iu,vRepresenting a set of programs rated by user u and neighbors v simultaneously, IuAnd IvRespectively representing program sets which are rated by a user u and a neighbor v; r isu,iAnd rv,iRespectively representing the scores of user u and neighbor v for program i,
and
representing the average score of user u and neighbor v.
5. The method of claim 3, wherein the similarity of the content features between the target user's preference and the nearest neighbors is calculated by:
the similarity between the content features is calculated by means of the similarity of the text, wherein,
a content feature vector representing nearest neighbor broadcast program, U { (U)1,w1),(u2,w2),...,(un,wn) The user interest content feature vector is represented.
6. The method for transient computation-based television recommendation of claim 3 wherein predicting program scores based on total similarity comprises:
selecting k users most similar to the target user according to the total similarity sim; then, the score of the user u for the program i is predicted based on the most similar users:
where V is a set of k similar users, rv,iIs the rating of the item i by the user v,
and
represents the average score of u and v.
7. The transient-computation-based television recommendation method of claim 1, wherein the video recommendation model is transiently computed based on reinforcement learning, and wherein modifying the recommendation model comprises:
after each behavior data is obtained, instantaneous calculation is carried out on the video recommendation model once based on reinforcement learning, model correction is carried out, and a recommendation result is updated.
8. A television recommendation system based on transient computing, comprising:
the data acquisition module is used for acquiring article content data and user behavior data and constructing a content matrix and a user matrix; the user behavior data comprises service data, log data, buried point data and external data;
the data processing module is used for carrying out hierarchical text classification on the content matrix and establishing a knowledge map; performing family portrait modeling on a user matrix;
the recommendation model establishing module is used for establishing a recommendation model based on the content matrix and the user matrix;
the initial program recommending module is used for recommending the initial program to the user based on the current article content data according to the recommending model;
the instantaneous calculation module receives behavior data of a user aiming at the recommendation result, recalls the recommendation result based on an instantaneous calculation engine of reinforcement learning, corrects the recommendation model through calculation loss and updates the recommendation result; the behavior data includes all clicking behaviors performed by the user through a control device such as a remote controller.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method for tv recommendation based on instantaneous calculation according to any of claims 1-7 when executing the program.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the method for tv recommendation based on instantaneous calculation according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910672136.XA CN110430471B (en) | 2019-07-24 | 2019-07-24 | Television recommendation method and system based on instantaneous calculation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910672136.XA CN110430471B (en) | 2019-07-24 | 2019-07-24 | Television recommendation method and system based on instantaneous calculation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110430471A CN110430471A (en) | 2019-11-08 |
| CN110430471B true CN110430471B (en) | 2021-05-07 |
Family
ID=68412130
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910672136.XA Active CN110430471B (en) | 2019-07-24 | 2019-07-24 | Television recommendation method and system based on instantaneous calculation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110430471B (en) |
Families Citing this family (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110908575B (en) * | 2019-12-05 | 2021-05-04 | 上海斑马来拉物流科技有限公司 | Data processing method, computer storage medium and electronic device |
| CN111858972B (en) * | 2020-07-28 | 2023-01-31 | 山东大学 | Movie recommendation method based on family knowledge graph |
| CN113709570A (en) * | 2020-09-25 | 2021-11-26 | 天翼智慧家庭科技有限公司 | Apparatus and method for recommending bandwidth based on IPTV probe data |
| CN112312216B (en) * | 2020-10-16 | 2022-08-16 | 山东海看新媒体研究院有限公司 | Traceable television recommendation method and system based on modular factorial theory |
| CN114528469A (en) * | 2020-11-23 | 2022-05-24 | 中兴通讯股份有限公司 | Recommendation method and device, electronic equipment and storage medium |
| CN112468853B (en) * | 2020-11-26 | 2023-01-03 | 未来电视有限公司 | Television resource recommendation method and device, computer equipment and storage medium |
| CN112364203B (en) * | 2020-11-30 | 2023-04-28 | 未来电视有限公司 | Television video recommendation method, device, server and storage medium |
| CN112507104B (en) * | 2020-12-18 | 2022-07-22 | 北京百度网讯科技有限公司 | Dialog system acquisition method, apparatus, storage medium and computer program product |
| CN112770181A (en) * | 2021-01-12 | 2021-05-07 | 贵州省广播电视信息网络股份有限公司 | Quick verification system and method for recommended content of family group |
| CN113051468B (en) * | 2021-02-22 | 2023-04-07 | 山东师范大学 | Movie recommendation method and system based on knowledge graph and reinforcement learning |
| CN112925723B (en) * | 2021-04-02 | 2022-03-15 | 上海复深蓝软件股份有限公司 | Test service recommendation method and device, computer equipment and storage medium |
| CN113630631A (en) * | 2021-08-23 | 2021-11-09 | 南京金智视讯技术有限公司 | HLS caching method and system based on collaborative filtering recommendation algorithm |
| CN114025205A (en) * | 2021-11-02 | 2022-02-08 | 天津大学 | Intelligent recommendation method for home TV video |
| CN114915844B (en) * | 2021-11-08 | 2023-02-28 | 海看网络科技(山东)股份有限公司 | Method and system for realizing real-time intelligent recommendation on IPTV |
| CN114915800A (en) * | 2021-12-07 | 2022-08-16 | 天翼数字生活科技有限公司 | System and method for predicting age and gender distribution of IPTV (Internet protocol television) family users |
| CN116628829B (en) * | 2023-07-24 | 2023-11-07 | 山东融谷信息科技有限公司 | Intelligent building three-dimensional visualization system based on digital twinning |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103152618B (en) * | 2011-12-07 | 2017-11-17 | 北京四达时代软件技术股份有限公司 | Value added service of digital television content recommendation method and device |
| CN105430505B (en) * | 2015-11-13 | 2018-07-03 | 云南大学 | A kind of IPTV program commending methods based on combined strategy |
| CN107862532B (en) * | 2016-09-22 | 2021-11-26 | 腾讯科技（深圳）有限公司 | User feature extraction method and related device |
| CN107454474B (en) * | 2017-08-17 | 2019-11-05 | 四川长虹电器股份有限公司 | A kind of television terminal program personalized recommendation method based on collaborative filtering |
| CN108650532B (en) * | 2018-03-22 | 2020-06-12 | 中国传媒大学 | Cable television on-demand program recommendation method and system |
| CN109672938A (en) * | 2019-01-07 | 2019-04-23 | 河北工业大学 | A kind of IPTV program commending method |
-
2019
- 2019-07-24 CN CN201910672136.XA patent/CN110430471B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN110430471A (en) | 2019-11-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110430471B (en) | Television recommendation method and system based on instantaneous calculation | |
| CN111444428B (en) | Information recommendation method and device based on artificial intelligence, electronic equipment and storage medium | |
| CN106802956B (en) | Movie recommendation method based on weighted heterogeneous information network | |
| CN106156127B (en) | Method and device for selecting data content to push to terminal | |
| US8260117B1 (en) | Automatically recommending content | |
| US8620917B2 (en) | Symantic framework for dynamically creating a program guide | |
| CN108920577A (en) | Television set intelligently recommended method | |
| CN111861550A (en) | OTT (over the Top) equipment-based family portrait construction method and system | |
| CN108664558B (en) | Network television personalized recommendation service method for large-scale users | |
| CN111062527A (en) | Video collection flow prediction method and device | |
| WO2015025248A2 (en) | A system apparatus circuit method and associated computer executable code for hybrid content recommendation | |
| CN112380451A (en) | Favorite content recommendation method based on big data | |
| CN113051468B (en) | Movie recommendation method and system based on knowledge graph and reinforcement learning | |
| CN113836406A (en) | Information flow recommendation method and device | |
| CN109508407A (en) | The tv product recommended method of time of fusion and Interest Similarity | |
| CN114358807A (en) | User portrayal method and system based on predictable user characteristic attributes | |
| CN113449200B (en) | Article recommendation method and device and computer storage medium | |
| CN110083766B (en) | Query recommendation method and device based on meta-path guiding embedding | |
| CN113468413B (en) | Multi-user sharing-oriented multimedia network video recommendation method | |
| CN116010711A (en) | KGCN model movie recommendation method integrating user information and interest change | |
| CN115618101A (en) | Streaming media content recommendation method and device based on negative feedback and electronic equipment | |
| CN112312216B (en) | Traceable television recommendation method and system based on modular factorial theory | |
| CN114915844B (en) | Method and system for realizing real-time intelligent recommendation on IPTV | |
| CN114329167A (en) | Hyper-parameter learning, intelligent recommendation, keyword and multimedia recommendation method and device | |
| Sun et al. | Research of Personalized Recommendation Algorithm Based on Trust and User's Interest |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |