CN114201642A - Media content recommendation method and device, electronic equipment and storage medium - Google Patents
Media content recommendation method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN114201642A CN114201642A CN202111361991.2A CN202111361991A CN114201642A CN 114201642 A CN114201642 A CN 114201642A CN 202111361991 A CN202111361991 A CN 202111361991A CN 114201642 A CN114201642 A CN 114201642A
- Authority
- CN
- China
- Prior art keywords
- media content
- user account
- media
- preset
- correlation index
- 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
- 238000000034 method Methods 0.000 title claims abstract description 62
- 230000008569 process Effects 0.000 claims abstract description 16
- 230000000875 corresponding effect Effects 0.000 claims description 43
- 230000006399 behavior Effects 0.000 claims description 22
- 238000004590 computer program Methods 0.000 claims description 13
- 230000002596 correlated effect Effects 0.000 claims description 10
- 239000013585 weight reducing agent Substances 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 abstract description 7
- 238000004422 calculation algorithm Methods 0.000 description 19
- 230000006870 function Effects 0.000 description 11
- 238000001914 filtration Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 8
- 230000000694 effects Effects 0.000 description 5
- 230000004044 response Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 235000019633 pungent taste Nutrition 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/73—Querying
- G06F16/735—Filtering based on additional data, e.g. user or group profiles
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
Abstract
The present disclosure relates to a media content recommendation method, apparatus, electronic device and storage medium, the method comprising: responding to a media content recommendation request of a target user account, and determining preferred media content of the target user account and a plurality of first media contents corresponding to the preferred media content; the first media content refers to media content which is continuously operated by the user account in the process of operating the preferred media content by the user account; acquiring a correlation index between the preference media content and each first media content in the plurality of first media contents from a preset storage area; determining target media content to be recommended from the plurality of first media content based on the relevance index; the target media content is sent to the target user account. The method and the device can improve the recommendation accuracy of the media content and have low calculation complexity.
Description
Technical Field
The present disclosure relates to the field of computer technologies, and in particular, to a method and an apparatus for recommending media content, an electronic device, and a storage medium.
Background
The recommendation system is an information filtering system, and can learn interests and hobbies of a user and predict scores or preferences of the user on given articles according to files or historical behavior records of the user, so that the communication mode between a platform and the user is changed, the interactivity between the platform and the user is enhanced, and works which may be interested are recommended to the user. At present, the recommendation system is widely applied to the fields of e-commerce platforms, news software, music software, short videos and the like.
Taking a short-video recommendation system as an example, the recommendation process mainly comprises two stages of recall and sequencing, wherein the recall stage is to retrieve thousands of videos which are possibly interested by users from millions of video pools; and in the sorting stage, the recalled videos are sorted according to the interest preference of the user, and then a plurality of videos with the highest scores are pushed to the user.
Wherein, whether the content which is interested by the user can be recalled from the mass video pool in the recalling stage determines the upper limit of the recommendation effect. The recall algorithm used in the recall stage generally infers which videos the user is interested in by analyzing the user's historical behavioral data and then recommends similar videos to the user.
The collaborative filtering algorithm is the most classical and most applied one of the recall algorithms. The core point of the collaborative filtering algorithm is how to calculate the similarity between videos according to the historical behaviors of users. The accuracy of the video similarity is improved, and the recommendation effect can be improved.
However, currently, commonly used collaborative filtering algorithms such as common neighbors algorithm, adaptive/Adar algorithm and Swing algorithm, wherein common neighbors algorithm and adaptive/Adar algorithm have the advantages of low algorithm complexity, easy real-time training, high recall rate and poor accuracy; the Swing algorithm has the advantages of high accuracy and low recall rate, and has the disadvantages of high algorithm complexity, difficulty in real-time training and low recall rate; therefore, the relation between the accuracy and the calculation complexity is difficult to balance by the conventional collaborative filtering algorithm; in addition, the conventional collaborative filtering algorithm lacks a recall mode for personalized users, popular videos are easy to recommend, and personalized video recommendation cannot be provided for different users.
Disclosure of Invention
The present disclosure provides a method and an apparatus for recommending media content, an electronic device, and a storage medium, which can reduce the right of popular media content, improve the accuracy of recommendation, and have low complexity and good real-time effect. The technical scheme of the disclosure is as follows:
according to a first aspect of the embodiments of the present disclosure, there is provided a media content recommendation method, including:
responding to a media content recommendation request of a target user account, and determining preferred media content of the target user account and a plurality of first media contents corresponding to the preferred media content; the first media content refers to media content which is continuously operated by the user account in the process of operating the preferred media content by the user account;
acquiring a correlation index between the preference media content and each first media content in the plurality of first media contents from a preset storage area;
determining target media content to be recommended from the plurality of first media content based on the relevance index;
the target media content is sent to the target user account.
Optionally, the method further includes performing the following steps according to a preset cycle:
acquiring media contents operated by each user account accessing a recommendation system in a current period;
aiming at second media content and third media content which are continuously operated by the same user account in the current period: determining a correlation index between the second media content and the third media content according to historical operation data of a co-occurrence user account which has operated on the second media content and the third media content, user historical operation information corresponding to the third media content and the hot degree of the third media content; wherein the third media content is operated by the user account after the second media content;
and storing the correlation index between the second media content and the third media content in a preset storage area.
Optionally, the historical operation data of the co-occurrence user account includes the number of media contents operated by the co-occurrence user account within a preset historical time period;
determining a correlation index between the second media content and the third media content according to historical operation data of a co-occurrence user account having an operation on the second media content and the third media content, user historical operation information corresponding to the third media content, and the hot degree of the third media content, wherein the determining includes:
acquiring the number of first user accounts which have accessed the recommendation system in a preset historical time period and the number of second user accounts which operate third media contents in the preset historical time period;
determining a first parameter value according to the number of the first user accounts and the number of the second user accounts; the first parameter value characterizes a trending degree of the third media content;
determining a second parameter value according to the number of the media contents operated by each co-occurrence user account in a preset historical time period;
carrying out weighted summation on different types of operation behaviors in the user historical operation information of the third media content to obtain a third parameter value;
a relevance indicator between the second media content and the third media content is determined based on the first parameter value, the second parameter value and the third parameter value.
Optionally, the correlation index between the second media content and the third media content is negatively correlated with the second parameter value, and the correlation index between the second media content and the third media content is positively correlated with the first parameter value and the third parameter value, respectively.
Optionally, when the relevance index between the second media content and the third media content determined in the previous period of the current period is the first relevance index, the method further includes:
if the correlation index between the second media content and the third media content is determined to be the second correlation index in the current period, replacing the first correlation index with the second correlation index;
or;
and if the correlation index between the second media content and the third media content does not exist in the current period, performing weight reduction on the first correlation index in an exponential decay mode to obtain an updated first correlation index.
Optionally, when the correlation index between the second media content and the third media content in the preset storage area is lower than the preset value, the correlation index lower than the preset value is deleted.
Optionally, determining the target media content to be recommended from the plurality of first media contents based on the relevance index includes:
ranking the relevance indexes between the preference media content and each first media content, and determining the first media content meeting the preset conditions as target media content to be recommended;
the preset condition comprises that the relevance index is larger than or equal to a threshold value or the top N ranking digits.
Optionally, the preferred media content refers to media content that has been subjected to forward operation by the target user account;
the forward operation comprises any one or more of a praise operation, a focus operation, a comment operation and a play operation with the play time length being larger than the preset time length.
According to a second aspect of the embodiments of the present disclosure, there is provided a media content recommendation apparatus including:
the first determination module is configured to execute a media content recommendation request responding to a target user account, and determine preferred media content of the target user account and a plurality of first media contents corresponding to the preferred media content; the first media content refers to media content which is continuously operated by the user account in the process of operating the preferred media content by the user account;
an obtaining module configured to perform obtaining of a correlation index between the preferred media content and each of the plurality of first media contents from a preset storage area;
a second determination module configured to perform determining a target media content to be recommended from the plurality of first media contents based on the relevance indicator;
a sending module configured to perform sending the target media content to the target user account.
Optionally, the apparatus further comprises a circulation module;
a loop module configured to perform the following steps at preset periods: acquiring media contents operated by each user account accessing a recommendation system in a current period; aiming at second media content and third media content which are continuously operated by the same user account in the current period: determining a correlation index between the second media content and the third media content according to historical operation data of a co-occurrence user account which has operated on the second media content and the third media content, user historical operation information corresponding to the third media content and the hot degree of the third media content; wherein the third media content is operated by the user account after the second media content; and storing the correlation index between the second media content and the third media content in a preset storage area.
Optionally, the historical operation data of the co-occurrence user account includes the number of media contents operated by the co-occurrence user account within a preset historical time period;
a loop module configured to perform obtaining a first number of user accounts that have accessed the recommendation system within a preset historical period of time and a second number of user accounts that have operated the third media content within the preset historical period of time; determining a first parameter value according to the number of the first user accounts and the number of the second user accounts; the first parameter value characterizes a trending degree of the third media content; determining a second parameter value according to the number of the media contents operated by each co-occurrence user account in a preset historical time period; carrying out weighted summation on different types of operation behaviors in the user historical operation information of the third media content to obtain a third parameter value; a relevance indicator between the second media content and the third media content is determined based on the first parameter value, the second parameter value and the third parameter value.
Optionally, the correlation index between the second media content and the third media content is negatively correlated to the second parameter value, and the correlation index between the second media content and the third media content is positively correlated to the first parameter value and the third parameter value, respectively.
Optionally, when the relevance indicator between the second media content and the third media content determined in the previous period of the current period is the first relevance indicator, the apparatus further includes:
a replacement module configured to perform, if it is determined within the current period that the correlation index between the second media content and the third media content is the second correlation index, replacing the first correlation index with the second correlation index;
or;
and the updating module is configured to execute the step of reducing the weight of the first relevance index in an exponential decay mode to obtain the updated first relevance index if the relevance index between the second media content and the third media content does not exist in the current period.
Optionally, the apparatus further comprises:
and the deleting module is configured to delete the correlation index which is lower than the preset value when the correlation index between the second media content and the third media content in the preset storage area is lower than the preset value.
Optionally, the second determining module is configured to perform ranking on the relevance indexes between the preferred media content and each of the first media content, and determine the first media content meeting the preset condition as the target media content to be recommended; the preset condition comprises that the relevance index is larger than or equal to a threshold value or the top N ranking digits.
Optionally, the preferred media content refers to media content that has been subjected to forward operation by the target user account;
the forward operation comprises any one or more of a praise operation, a focus operation, a comment operation and a play operation with the play time length being larger than the preset time length.
According to a third aspect of the embodiments of the present disclosure, there is provided an electronic apparatus including:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to execute the instructions to implement the media content recommendation method of the first aspect.
According to a fourth aspect of embodiments of the present disclosure, there is provided a computer-readable storage medium, wherein instructions of the computer-readable storage medium, when executed by a processor of an electronic device, enable the electronic device to perform the media content recommendation method of the first aspect.
According to a fifth aspect of embodiments of the present disclosure, there is provided a computer program product, the computer program product comprising a computer program, the computer program being stored in a readable storage medium, from which at least one processor of a computer device reads and executes the computer program, so that the computer device performs the media content recommendation method of the first aspect described above.
The technical scheme provided by the embodiment of the disclosure at least brings the following beneficial effects:
determining preferred media content of a target user account and a plurality of first media contents corresponding to the preferred media content by responding to a media content recommendation request of the target user account; the first media content refers to media content which is continuously operated by the user account in the process of operating the preferred media content by the user account; acquiring a correlation index between the preference media content and each first media content in the plurality of first media contents from a preset storage area; determining target media content to be recommended from the plurality of first media content based on the relevance index; the target media content is sent to the target user account. In the technical scheme, the relevance indexes between the preference media content and the first media contents are calculated in advance and stored in the preset storage area, so that when a content recommendation request of a target user account is received, the relevance indexes can be directly obtained from the preset storage area, the target media contents are determined based on the relevance indexes, the recommendation efficiency can be improved, and the real-time effect is good; and because the first media content is the media content which is continuously operated by the user account in the process of operating the preferred media content by the user account, namely, the first media content has stronger time correlation with the preferred media content, and the content similarity between the continuously operated media content in time is higher based on big data analysis, the similarity between the target media content recommended to the target user account by the method and the device is higher, the recommended media content can better accord with the preference of the target user account, and the accuracy of media content recommendation can be improved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the disclosure and are not to be construed as limiting the disclosure.
FIG. 1 is a schematic diagram of an application environment illustrating a method of media content recommendation, according to an exemplary embodiment;
FIG. 2 is a flow diagram illustrating a method of media content recommendation, according to an exemplary embodiment;
FIG. 3 is a flow diagram illustrating the determination of a relevance indicator, according to an exemplary embodiment;
FIG. 4 is a flowchart illustrating a method of determining a relevance indicator between second media content and third media content, according to an example embodiment;
FIG. 5 is a block diagram illustrating a media content recommender, in accordance with an exemplary embodiment;
FIG. 6 is a block diagram illustrating an electronic device for media content recommendation, according to an example embodiment.
Detailed Description
In order to make the technical solutions of the present disclosure better understood by those of ordinary skill in the art, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings.
It should be noted that the terms "first," "second," and the like in the description and claims of the present disclosure and in the above-described drawings are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the disclosure described herein are capable of operation in sequences other than those illustrated or otherwise described herein. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.
In the field of short video recommendation, a commonly used collaborative filtering algorithm usually calculates the number of co-occurrence users of two videos when calculating the similarity between the two videos, wherein the larger the number of co-occurrence users is, the higher the calculated similarity is, and the number of co-occurrence users refers to a user set having an operation (clicking, watching, etc.) on the two videos. The existing collaborative filtering algorithm has the problems that high accuracy cannot be achieved under the condition of ensuring the calculation complexity, and the calculation complexity is higher under the condition of ensuring the accuracy, namely, the relation between the accuracy and the calculation complexity is difficult to balance; on the other hand, when the existing collaborative filtering algorithm recommends short videos to users, the popular videos tend to be recommended, and the contents pushed by the system are uniform over time, which is easy to cause aesthetic fatigue. Therefore, the recommendation system needs to mine interest points of different users, recommend content that the users are more likely to like to the users, and broaden the content broadness.
In view of this, the present disclosure provides a media content recommendation method, which, in response to a media content recommendation request of a target user account, first determines a preferred media content of the target user account, then determines a plurality of first media contents corresponding to the preferred media content, and determines a target media content recommended to the target user account based on a similarity index pre-calculated between each first media content and the preferred media content. The first media content refers to media content which is continuously operated by a user account in the process of operating the preferred media content by the user account, namely, stronger time correlation exists between the first media content and the preferred media content, and the content similarity between the continuously operated media content in time is higher based on big data analysis, so that the similarity between the target media content recommended to the target user account by the method and the device is higher, the recommended media content can better accord with the preference of the target user account, and the accuracy of media content recommendation can be improved. In addition, the similarity index is pre-calculated and stored in the preset storage area, so that when a recommendation request of a target user account is received, the similarity index can be directly called to determine the target media content, and therefore the calculation complexity of the recommendation process can be reduced, and the real-time effect is good.
Referring to fig. 1, a schematic diagram of an application environment of a media content recommendation method according to an exemplary embodiment is shown, where the application environment may include a terminal 110 and a server 120, and the terminal 110 and the server 120 may be connected through a wired network or a wireless network.
The terminal 110 may be, but is not limited to, a smart phone, a tablet computer, a notebook computer, a desktop computer, and the like. The terminal 110 may have client software such as an Application (App) installed therein, and the Application may be a stand-alone Application or a sub-program in the Application. Illustratively, the application may be a news-type application, a live-type application, or a video-type application, among others. The user of the terminal 110 may log into the application through pre-registered user information, which may include an account number and a password.
The server 120 may be a server providing background services for the application program in the terminal 110, and specifically, the service provided by the server 120 may be a media content recommendation service, and the media content may be determined according to a specific application scenario, and may include, but is not limited to, short video, news information, advertisement, and so on. The server 120 may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing basic cloud computing services such as a cloud service, a cloud database, cloud computing, a cloud function, cloud storage, a network service, cloud communication, middleware service, a domain name service, a security service, a CDN, a big data and artificial intelligence platform, and the like.
In one particular application scenario, the terminal 110 may send a media content recommendation request to the server 120 in response to a media content recommendation instruction, where the media content recommendation instruction may be generated based on a refresh operation when a user account (i.e., a target user account) of the terminal 110 refreshes a current page.
After receiving the media content recommendation request from the target user account, the server 120 may obtain the preferred media content of the target user account and determine a plurality of first media contents corresponding to the preferred media content; the first media content refers to media content which is continuously operated by any user account in the process of operating preferred media content by the user account; here, any user account may be the target user account, or may be another user account accessing the recommendation system; then, the server 120 obtains the correlation index between the preferred media content and each first media content in the plurality of first media contents from the preset storage area, where the correlation index between the preferred media content and each first media content is pre-calculated, and therefore, the correlation index can be directly called from the preset storage area; then, the server 120 determines a target media content to be recommended from the plurality of first media contents based on the relevance index, and sends the target media content to the terminal 110 corresponding to the target user account, so that the terminal 110 displays the target media content.
It should be understood that the application environment shown in fig. 1 is only an example, and in practical applications, the terminal or the server may independently execute the interest probability determination method according to the embodiment of the present disclosure, or the terminal and the server may cooperatively execute the interest probability determination method according to the embodiment of the present disclosure, and the embodiment of the present disclosure is not limited to a specific application environment.
In the embodiment of the disclosure, the server may receive media content recommendation requests of a plurality of user accounts logged in a plurality of terminals, and perform media content recommendation for each user account in a personalized manner for the user account according to the media content recommendation request of each user account. In the following, a user account of a plurality of user accounts is taken as a target user account for explanation, a specific recommendation method is shown in fig. 2, fig. 2 is a flowchart of a media content recommendation method according to an exemplary embodiment, taking the case that the media content recommendation method is used in the server in fig. 1 as an example, the method includes the following steps:
in step S201, in response to the media content recommendation request of the target user account, the preferred media content of the target user account and a plurality of first media contents corresponding to the preferred media content are determined.
The first media content refers to media content which is continuously operated by the user account in the process of operating the preferred media content by the user account.
In the embodiment of the present disclosure, the media content may include video, picture, advertisement, news, and other content that can be distributed over the internet. The target user account may trigger generation of a media content recommendation instruction based on a preset operation in a preset application program of the terminal, and the terminal sends a media content recommendation request to the recommendation server based on the media content recommendation instruction, where the media content recommendation request may carry an identifier of the target user account. And the server responds to the media content recommendation request and determines the preferred media content of the target user account according to the identification of the target user account.
It should be noted that, in the embodiment of the present disclosure, the server may provide a recommendation service for each user account accessing the recommendation system, and the server needs to collect a behavior log of each user account, where the behavior log includes media content operation data of each user account, and recommend media content for a target user account based on the media content operation data, and therefore, the target user account referred to herein refers to a user account accessing the recommendation system and sending a media content recommendation request to the server.
In an optional implementation manner, the preferred media content refers to media content that has been performed forward operation by the target user account; the forward operation comprises any one or more of a praise operation, a focus operation, a comment operation and a play operation with the play time length being larger than the preset time length. Taking the short video application scenario as an example, the preferred media content may include a short video that is approved by the target user account, and may also include a short video that is commented on by the target user account.
In an alternative embodiment, the number of the preferred media contents may be determined according to actual requirements, for example, a preset number of preferred media contents may be randomly acquired, where the number of media contents on which a forward operation is performed by the target user account in the preset number of preferred media contents is a random number, for example, 200 preferred videos are randomly acquired, where the number of videos on which a favorite operation is performed by the target user account in the 200 preferred videos is random, and the number of videos on which a comment operation is performed by the target user account is also random; for another example, for different types of forward operations, a preset number of preferred media contents corresponding to each type of forward operation may be acquired, such as 50 videos for which the target user account has performed a praise operation recently, 50 videos for which a care operation has been performed, 50 videos for which a comment operation has been performed, and 50 videos for which the play duration is greater than 20 s.
When the number of the preferred media contents is plural, in the step S201, it is necessary to determine a plurality of corresponding first media contents for each preferred media content.
In the embodiment of the present disclosure, the first media content refers to media content that is continuously operated by a user account during a process of the user account operating preferred media content, where the user account refers to any user account accessing a recommendation system, and it can be understood that the user account includes a target user account and also includes other user accounts that have operated preferred media content and the first media content. The first media content and the preferred media content are operated by the same user account continuously in time, and the first media content can be the media content which is operated continuously after the user account operates the preferred media content first; the first media content may also be media content operated by the user account before the preferred media content is operated, in other words, the first media content may be operated by the user account first, and the preferred media content is media content continuously operated by the user account after the first media content; for example, the preferred media content of the target user account a is a video x, and the first media content corresponding to the video x includes a video y, where the video y is a video that the user account B continuously watches by the user account B during watching the video x, that is, the user account B watches the video x first and then watches the video y again; or, the user account B watches the video y first, and then watches the video x next; in both cases, video y may be referred to as the first media content of video x.
The recommendation method of the embodiment of the disclosure determines preferred media content of the target user account based on a media content recommendation request of the target user account, and recommends the media content on the basis of the preferred media content, wherein the preferred media content is media content which is subjected to forward operations such as a praise operation, an attention operation, a comment operation and a play operation with a play time longer than a preset time length and the like by the target user account, so that the calculation accuracy of the target media content can be improved, and the media content recommended to the target user account is more in line with the preference of the target user account; in addition, the target media content is derived from a plurality of first media contents corresponding to the preference media content, and due to the fact that strong time correlation exists between the first media contents and the preference media content, the content similarity between the first media contents and the preference media content is often higher in the time correlation compared with that between the common media contents and the preference media content, and therefore the accuracy of media content recommendation can be further improved.
In step S203, correlation indexes between the preferred media content and each of the plurality of first media contents are acquired from a preset storage area.
In the embodiment of the disclosure, the correlation index between the preferred media content and each of the plurality of first media contents is pre-calculated and stored in the preset storage area, so that the server can directly obtain the plurality of first media contents corresponding to the preferred media content and the correlation index between the preferred media content and each of the plurality of first media contents from the preset storage area in response to the media content recommendation request of the target user account.
In an alternative embodiment, the correlation index may be calculated in the following steps S301 to S303. As shown in fig. 3, the method of the embodiment of the present disclosure may further include: executing the following steps according to a preset period:
in step S301, media content that has been operated in the current period by each user account accessing the recommendation system is acquired.
In the step, the server logs the behavior of each user account of the recommendation system in real time, and obtains the media content operated by each user account in the current period at preset time intervals. Taking the media content as a video as an example, the behavior log comprises the identification of a user account, the identification of each video operated by the user account, the time length of playing the video by the user account, whether the user account approves, pays attention to and comments on the user account, and the like; the preset time interval may be set according to actual requirements, and is characterized by a preset time length, for example, the preset time interval may be 5 minutes.
In step S303, for the second media content and the third media content that are continuously operated by the same user account in the current period: determining a correlation index between the second media content and the third media content according to historical operation data of a co-occurrence user account which has operated on the second media content and the third media content, user historical operation information corresponding to the third media content and the hot degree of the third media content; wherein the third media content is operated by the user account after the second media content.
In step S305, a correlation index between the second media content and the third media content is stored in a preset storage area.
In the above steps, for the second media content and the third media content that are continuously operated by the same user account in the current period, that is, as long as there is a user account, which operates the second media content first and then operates the third media content, the correlation index between the second media content and the third media content is calculated. Taking the media content as an example, assuming that the user account accessing the recommendation system in the current period includes a, the server may acquire that the user account a plays videos p1, p2, p3 and p4 in 5 minutes, and at this time, correlation indexes between p1 and p2, between p2 and p3, and between p3 and p4 need to be calculated; for example, when the correlation index between p1 and p2 is calculated, p1 is the second media content, and p2 is the third media content.
In an optional embodiment, the historical operation data of the co-occurrence user account includes the number of media contents operated by the co-occurrence user account in a preset historical time period; the preset historical time period is characterized by a preset time length, for example, the preset historical time period is 3 days.
Correspondingly, in the step S303, the determining a correlation index between the second media content and the third media content according to the historical operation data of the co-occurrence user account having an operation on the second media content and the third media content, the user historical operation information corresponding to the third media content, and the trending degree of the third media content may specifically include the following steps as shown in fig. 4:
in step S401, a first user account number for which the recommendation system has been accessed within a preset history time period and a second user account number for which the third media content has been operated within the preset history time period are acquired.
In step S403, determining a first parameter value according to the first user account number and the second user account number; the first parameter value characterizes a degree of hotness of the third media content.
In this step, in order to implement the weight reduction of the trending media content and reduce the frequency of recommending the trending media content, the relevance index between the trending media content and other media content is reduced when the relevance index is calculated, so that by defining a first parameter value representing the trending degree of the third media content, the first parameter value is positively correlated with the number of second user accounts operating the third media content within the preset historical time period. Specifically, a ratio of the number of the first user accounts to the number of the second user accounts is subjected to preset function processing, and a value processed by the preset function is used as a first parameter value. For example, in a specific implementation, the predetermined function is a log function, and accordingly, the first parameter value is calculated as the following formula (1):
wherein, IDFyA first parameter value representing the third media content y; u represents the number of first user accounts which have accessed the recommendation system in a preset historical time period; Γ (y) represents a second number of user accounts operating on the third media content y for a preset historical period of time; wherein, the larger Γ (y) is, the hotter the third media content y is.
In the specific implementation, in the process of determining the relevance index, a concept of Inverse Document Frequency (IDF) is introduced, and the IDF is often used for representing the general importance of a word in the field of natural language processing. The IDF value for a particular term may be obtained by dividing the total number of documents by the number of documents that contain that term and taking the logarithm of the resulting quotient. If the number of documents containing the entry t is smaller, the IDF value of the entry t is larger, and the entry t has good category distinguishing capability. If the document number of the entry t in a certain class of document C is m, and the total number of documents of other classes containing t is k, it is obvious that the document number n of all the entries containing t is m + k, and when m is large, the total number n of the documents is also large, the IDF value of the entry t is small, which indicates that the category distinguishing capability of the entry t is not strong.
Similarly, as can be seen from the definition of equation (1) above, the numerator of the function is the first number of user accounts that have accessed the recommendation system, and the denominator is the second number of user accounts that have manipulated the third media content y, such that for a popular third media content y, the Γ (y) is larger, and the corresponding IDF is largeryWill be smaller. It can be seen that the more popular the third media content is, the smaller the first parameter value will be, which brings about the advantage that the less relevant index will be obtained by the popular third media content when the relevant index is calculated subsequently, which can realize the weight reduction of the popular content, and further can reduce the tendency of recommending the popular video, and is helpful for realizing the personalization of the recommendation service.
In step S405, a second parameter value is determined according to the number of media contents operated by each co-occurrence user account within a preset history time period.
The co-occurrence user accounts refer to user accounts which operate the second media content first and then operate the third media content within preset time, a plurality of co-occurrence user accounts may exist, for any co-occurrence user account, the larger the number of the media contents operated within the preset historical time period is, the co-occurrence user account belongs to a popular user, and since the popular user operates the media contents frequently and the referential performance of the operation data is low, when the subsequent correlation index is calculated, the second parameter value is taken as a negative correlation parameter, the right reduction of the popular user can be realized, and the negative influence of the operation data of the popular user on the accuracy of the correlation index can be avoided.
In step S407, the different types of operation behaviors in the user history operation information of the third media content are weighted and summed to obtain a third parameter value.
The user history operation information refers to specific operation information that is executed on the third media content by the co-occurrence user account within a preset history time period, and the specific operation information includes different types of operation behaviors. Taking media content as an example of a video, the different types of operation behaviors may include a play operation in which a video playing time exceeds a preset time, a praise operation, a focus operation, and a comment operation.
In a specific implementation, different weights may be set for the different types of operation behaviors, and a third parameter value may be obtained by fusing the operation behaviors of the co-occurrence user account on a third media content in a linear weighting manner, where the third parameter value may be defined as ΣiωyiactionyiWherein, actionyiRepresenting different types of operational behavior, ω, of any of the co-occurring user accounts for the third media content yyiAnd the weight of the corresponding operation behavior is represented as a hyperparameter. And obtaining a third parameter value corresponding to any co-occurrence user account by carrying out weighted summation on the operation behaviors of the third media content. Therefore, when the relevance index is calculated subsequently, the third parameter values corresponding to each co-occurrence user account are summed, the operation behaviors of the co-occurrence user accounts can be fused, and the accuracy of media content recommendation is improved.
Taking media content as video as an example, any co-occurrence user account for a certain videoThe different types of operation behaviors can include play operation actions with the video playing time length exceeding the preset time lengthy1Operation of clicking praisey2Focus operation actiony3And comment operation actiony4The weight corresponding to each operation may be ωy1=0.4,ωy2=0.2,ωy3=0.3,ωy40.1; then, the third parameter value corresponding to the co-occurrence user account is 0.4 actiony1+0.2*actiony2+0.3*actiony3+0.1*actiony4。
In step S409, a relevance indicator between the second media content and the third media content is determined based on the first parameter value, the second parameter value and the third parameter value.
In a particular implementation, the relevancy indicator between the second media content and the third media content is inversely related to the second parameter value. The correlation index between the second media content and the third media content is positively correlated with the first parameter value and the third parameter value respectively. In practical application, thousands of millions of co-occurrence user accounts may exist between the second media content and the third media content, and the value processed by the preset function is used as the second parameter value corresponding to each co-occurrence user account by performing the preset function processing on the number of the media contents operated by each co-occurrence user account in the preset historical time period. And when the correlation index is calculated, taking the reciprocal of the second parameter value corresponding to each co-occurrence user account, multiplying the reciprocal by the third parameter value corresponding to the co-occurrence user account, namely taking the third parameter value corresponding to the co-occurrence user account as a numerator and the second parameter value corresponding to the co-occurrence user account as a denominator to obtain the score value corresponding to the co-occurrence user account, summing the score values of all co-occurrence user numbers, and multiplying the sum by the first parameter value to obtain the correlation index. Wherein the preset function may be a log function; for example, the correlation index may be determined according to the following formula (2):
wherein, SimScorexyA relevance indicator representing the second media content x and the third media content y; u represents a co-occurrence user account operating on both the second media content x and the third media content y; IDFyRepresenting a first parameter value; phi (u) represents the number of media contents operated by the co-occurrence user account u in a preset historical time period, and log (| phi (u) |) represents a second parameter value corresponding to the co-occurrence user account processed by a preset function; sigmaiωyiactionyiA third parameter value, action, corresponding to a co-occurrence user accountyiRepresenting different types of operational behaviour, ω, of a contributing user account with respect to the third media content yyiAnd the weight of the corresponding operation behavior is represented as a hyperparameter.
In an optional implementation manner, when the relevance indicator between the second media content and the third media content determined in the previous cycle of the current cycle is the first relevance indicator, the method of the embodiment of the present disclosure further includes:
if the correlation index between the second media content and the third media content is determined to be the second correlation index in the current period, replacing the first correlation index with the second correlation index;
or;
and if the correlation index between the second media content and the third media content does not exist in the current period, performing weight reduction on the first correlation index in an exponential decay mode to obtain an updated first correlation index.
For example, in the first case, within 5 minutes of the last cycle of the current cycle, the first correlation index between the videos p1 and p2 is determined to be 0.9, and within 5 minutes of the current cycle, the second correlation index between p1 and p2 is determined to be 0.8, and then 0.9 is replaced by 0.8, so that the replacement of the correlation indexes is completed; in the second case, the first correlation index between p1 and p2 is calculated to be 0.9 in 5 minutes in the previous period, and if the correlation index between p1 and p2 does not exist in 5 minutes in the current period, the weight of 0.9 is reduced in an exponential decay manner to obtain an updated first correlation index, where the decay factor adopted in the exponential decay may be 0.95.
In a further optional embodiment, when the correlation index between the second media content and the third media content in the preset storage area is lower than a preset value, the correlation index lower than the preset value is deleted.
Wherein the preset value may be 0.1; therefore, if the weight of a certain correlation index is reduced by exponential decay for multiple times, the correlation index is finally deleted, so that the storage space of the preset storage area is saved.
Therefore, according to the method and the device, the relevance index is determined periodically and updated, so that the relevance index stored in the preset storage area has good timeliness, and more accurate media content can be recommended to the target user account in the follow-up process.
In step S205, a target media content to be recommended is determined from the plurality of first media contents based on the relevance index.
In step S207, the target media content is sent to the target user account.
In the embodiment of the present disclosure, the correlation index between the preferred media content and the first media content is calculated in advance in the above steps S301 to S303, and based on the foregoing description, the first media content may be the above second media content or the above third media content. When recommending, directly obtaining a plurality of first media contents corresponding to the preference media contents from the preset storage area, and the correlation indexes between the preference media contents and the first media contents.
In an optional implementation manner, the determining, by the foregoing step, a target media content to be recommended from a plurality of first media contents based on the relevance indicator may specifically include:
ranking the relevance indexes between the preference media content and each first media content, and determining the first media content meeting the preset conditions as target media content to be recommended;
the preset condition comprises that the relevance index is larger than or equal to a threshold value or the top N ranking digits.
In practical application, the number of the first media contents determined based on the preference media contents is thousands of, so that the first media contents in the thousands of first media contents need to be screened based on the relevance indexes corresponding to the first media contents, and the first media contents meeting the preset conditions are taken as target media contents to be recommended. The preset condition may include a fixed threshold or a dynamically adjustable threshold, and the dynamically adjustable threshold may be calculated according to the correlation index corresponding to each first media content. Or, the preset condition may be that the relevance indexes are ranked from large to small and then located at the top N ranking positions, for example, after ranking the relevance indexes corresponding to the first media contents, the first media contents corresponding to the relevance indexes of the top 1000 ranking positions are recommended to the target user account as the target media contents.
Compared with videos recommended by the existing algorithm, the candidate videos recommended to the target user account by the method are improved in playing time by 3.2%, the complimentary rate is improved by 1.4%, the attention rate is improved by 4.6%, the comment rate is improved by 2.1%, and the exposure proportion of hot videos is reduced by 8.4%.
FIG. 5 is a block diagram illustrating a media content recommender, according to an exemplary embodiment. Referring to fig. 5, the media content recommender 500 includes a first determining module 510, an obtaining module 520, a second determining module 530 and a sending module 540, wherein:
a first determining module 510 configured to execute determining, in response to a media content recommendation request of a target user account, preferred media content of the target user account and a plurality of first media contents corresponding to the preferred media content; the first media content refers to media content which is continuously operated by the user account in the process of operating the preferred media content by the user account;
an obtaining module 520 configured to perform obtaining of a correlation index between the preferred media content and each of the plurality of first media contents from a preset storage area;
a second determining module 530 configured to perform determining a target media content to be recommended from the plurality of first media contents based on the relevance indicator;
a sending module 540 configured to perform sending the target media content to the target user account.
In an alternative embodiment, the apparatus further comprises a circulation module;
a loop module configured to perform the following steps at preset periods: acquiring media contents operated by each user account accessing a recommendation system in a current period; aiming at second media content and third media content which are continuously operated by the same user account in the current period: determining a correlation index between the second media content and the third media content according to historical operation data of a co-occurrence user account which has operated on the second media content and the third media content, user historical operation information corresponding to the third media content and the hot degree of the third media content; wherein the third media content is operated by the user account after the second media content; and storing the correlation index between the second media content and the third media content in a preset storage area.
In an alternative embodiment, the historical operation data of the co-occurrence user account includes the number of media contents operated by the co-occurrence user account in a preset historical time period;
a loop module configured to perform obtaining a first number of user accounts that have accessed the recommendation system within a preset historical period of time and a second number of user accounts that have operated the third media content within the preset historical period of time; determining a first parameter value according to the number of the first user accounts and the number of the second user accounts; the first parameter value characterizes a trending degree of the third media content; determining a second parameter value according to the number of the media contents operated by each co-occurrence user account in a preset historical time period; carrying out weighted summation on different types of operation behaviors in the user historical operation information of the third media content to obtain a third parameter value; a relevance indicator between the second media content and the third media content is determined based on the first parameter value, the second parameter value and the third parameter value.
In an optional implementation manner, the correlation index between the second media content and the third media content is negatively correlated with the second parameter value, and the correlation index between the second media content and the third media content is positively correlated with the first parameter value and the third parameter value, respectively.
In an optional implementation manner, when the relevance indicator between the second media content and the third media content determined in the previous period of the current period is the first relevance indicator, the apparatus further includes:
a replacement module configured to perform, if it is determined within the current period that the correlation index between the second media content and the third media content is the second correlation index, replacing the first correlation index with the second correlation index;
or;
and the updating module is configured to execute the step of reducing the weight of the first relevance index in an exponential decay mode to obtain the updated first relevance index if the relevance index between the second media content and the third media content does not exist in the current period.
In an alternative embodiment, the apparatus further comprises:
and the deleting module is configured to delete the correlation index which is lower than the preset value when the correlation index between the second media content and the third media content in the preset storage area is lower than the preset value.
In an optional embodiment, the second determining module 530 is configured to perform ranking on the relevance indicators between the preferred media content and each first media content, and determine the first media content meeting a preset condition as the target media content to be recommended; the preset condition comprises that the relevance index is larger than or equal to a threshold value or the top N ranking digits.
In an alternative embodiment, the preferred media content refers to media content that has been subjected to forward operation by the target user account;
the forward operation comprises any one or more of a praise operation, a focus operation, a comment operation and a play operation with the play time length being larger than the preset time length.
With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.
In one exemplary embodiment, there is also provided an electronic device, comprising a processor; a memory for storing processor-executable instructions; wherein the processor is configured to execute the instructions stored in the memory to implement any one of the media content recommendation methods provided in the embodiments of the present disclosure.
The electronic device may be a terminal, a server, or a similar computing device, taking the electronic device as a server as an example, fig. 6 is a block diagram of an electronic device for media content recommendation according to an exemplary embodiment, and as shown in fig. 6, the server 600 may generate a relatively large difference due to different configurations or performances, and may include one or more Central Processing Units (CPUs) 610 (the processors 610 may include but are not limited to Processing devices such as a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 630 for storing data, and one or more storage media 620 (e.g., one or more mass storage devices) for storing an application program 623 or data 622. Memory 630 and storage medium 620 may be, among other things, transient or persistent storage. The program stored on the storage medium 620 may include one or more modules, each of which may include a series of instruction operations for the server. Still further, the central processor 610 may be configured to communicate with the storage medium 620 to execute a series of instruction operations in the storage medium 620 on the server 600. The server 600 may also include one or more power supplies 660, one or more wired or wireless network interfaces 650, one or more input-output interfaces 640, and/or one or more operating systems 621, such as Windows Server, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, and so forth.
The input/output interface 640 may be used to receive or transmit data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider of the server 600. In one example, i/o Interface 640 includes a Network adapter (NIC) that may be coupled to other Network devices via a base station to communicate with the internet. In one example, the input/output interface 640 may be a Radio Frequency (RF) module, which is used for communicating with the internet in a wireless manner.
It will be understood by those skilled in the art that the structure shown in fig. 6 is only an illustration and is not intended to limit the structure of the electronic device. For example, server 600 may also include more or fewer components than shown in FIG. 6, or have a different configuration than shown in FIG. 6.
In an exemplary embodiment, a computer-readable storage medium comprising instructions, such as the memory 630 comprising instructions, executable by the processor 610 of the apparatus 600 to perform the method described above is also provided. Alternatively, the computer readable storage medium may be a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.
In an exemplary embodiment, a computer program product is also provided, which includes a computer program stored in a readable storage medium, and at least one processor of a computer device reads and executes the computer program from the readable storage medium, so that the computer device executes any one of the media content recommendation methods provided in the embodiments of the present disclosure.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims (10)
1. A method for recommending media contents, comprising:
responding to a media content recommendation request of a target user account, and determining preferred media content of the target user account and a plurality of first media contents corresponding to the preferred media content; the first media content refers to media content which is continuously operated by a user account in the process of operating preferred media content by the user account;
acquiring a correlation index between the preferred media content and each first media content in the plurality of first media contents from a preset storage area;
determining target media content to be recommended from the plurality of first media content based on the relevance indicator;
sending the target media content to the target user account.
2. The media content recommendation method according to claim 1, further comprising the steps of, according to a preset period:
acquiring media content operated by each user account accessing a recommendation system in a current period;
aiming at second media content and third media content which are continuously operated by the same user account in the current period: determining a correlation index between the second media content and the third media content according to historical operation data of a co-occurrence user account having an operation on the second media content and the third media content, user historical operation information corresponding to the third media content and the trending degree of the third media content; wherein the third media content is operated by the user account after the second media content;
and storing the correlation index between the second media content and the third media content in the preset storage area.
3. The media content recommendation method according to claim 2, wherein the historical operation data of the co-occurrence user account comprises the number of media contents operated by the co-occurrence user account in a preset historical time period;
the determining a correlation index between the second media content and the third media content according to historical operation data of a co-occurrence user account having an operation on the second media content and the third media content, user historical operation information corresponding to the third media content, and a trending degree of the third media content includes:
acquiring the number of first user accounts which have accessed the recommendation system in the preset historical time period and the number of second user accounts which have operated the third media content in the preset historical time period;
determining a first parameter value according to the first user account number and the second user account number; the first parameter value characterizes a degree of trending of the third media content;
determining a second parameter value according to the number of the media contents operated by each co-occurrence user account in the preset historical time period;
carrying out weighted summation on different types of operation behaviors in the user historical operation information of the third media content to obtain a third parameter value;
determining a relevance indicator between the second media content and the third media content based on the first parameter value, the second parameter value, and the third parameter value.
4. The method of claim 3, wherein the correlation indicator between the second media content and the third media content is negatively correlated to the second parameter value, and the correlation indicator between the second media content and the third media content is positively correlated to the first parameter value and the third parameter value, respectively.
5. The media content recommendation method according to claim 2 or 3, wherein when the relevance index between the second media content and the third media content determined in the previous cycle of the current cycle is the first relevance index, the method further comprises:
if the correlation index between the second media content and the third media content is determined to be a second correlation index in the current period, replacing the first correlation index with the second correlation index;
or;
and if the correlation index between the second media content and the third media content does not exist in the current period, performing weight reduction on the first correlation index in an exponential decay mode to obtain an updated first correlation index.
6. The media content recommendation method of claim 5, further comprising:
and when the correlation index between the second media content and the third media content in the preset storage area is lower than a preset value, deleting the correlation index lower than the preset value.
7. A media content recommender, comprising:
the first determination module is configured to execute a media content recommendation request responding to a target user account, and determine preferred media content of the target user account and a plurality of first media contents corresponding to the preferred media content; the first media content refers to media content which is continuously operated by a user account in the process of operating preferred media content by the user account;
an obtaining module configured to perform obtaining of a correlation index between the preferred media content and each of the plurality of first media contents from a preset storage area;
a second determination module configured to perform determining target media content to be recommended from the plurality of first media content based on the relevance indicator;
a sending module configured to perform sending the target media content to the target user account.
8. An electronic device, comprising:
a processor;
a memory for storing the processor-executable instructions;
wherein the processor is configured to execute the instructions to implement the media content recommendation method of any of claims 1-6.
9. A computer-readable storage medium, wherein instructions in the computer-readable storage medium, when executed by a processor of an electronic device, enable the electronic device to perform the media content recommendation method of any of claims 1-6.
10. A computer program product, characterized in that the computer program product comprises a computer program, the computer program being stored in a readable storage medium, from which at least one processor of a computer device reads and executes the computer program, causing the computer device to perform the media content recommendation method according to any one of claims 1 to 6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111361991.2A CN114201642A (en) | 2021-11-17 | 2021-11-17 | Media content recommendation method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111361991.2A CN114201642A (en) | 2021-11-17 | 2021-11-17 | Media content recommendation method and device, electronic equipment and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114201642A true CN114201642A (en) | 2022-03-18 |
Family
ID=80647832
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111361991.2A Pending CN114201642A (en) | 2021-11-17 | 2021-11-17 | Media content recommendation method and device, electronic equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114201642A (en) |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108614856A (en) * | 2018-03-21 | 2018-10-02 | 北京奇艺世纪科技有限公司 | A kind of video sequence calibration method and device |
CN108804492A (en) * | 2018-03-27 | 2018-11-13 | 优视科技新加坡有限公司 | The method and device recommended for multimedia object |
US20190297365A1 (en) * | 2018-03-23 | 2019-09-26 | Rovi Guides, Inc. | Systems and methods for increasing a likelihood of media asset consumption using recommendation chains |
US20200021894A1 (en) * | 2018-07-12 | 2020-01-16 | Rovi Guides, Inc. | Systems and methods for gamification of real-time instructional commentating |
CN110781321A (en) * | 2019-08-28 | 2020-02-11 | 腾讯科技(深圳)有限公司 | Multimedia content recommendation method and device |
CN111708948A (en) * | 2020-06-19 | 2020-09-25 | 北京达佳互联信息技术有限公司 | Content item recommendation method, device, server and computer readable storage medium |
CN112364186A (en) * | 2020-11-30 | 2021-02-12 | 腾讯科技(北京)有限公司 | Method, device and equipment for presenting media recommendation information and storage medium |
CN113360744A (en) * | 2020-03-04 | 2021-09-07 | 北京达佳互联信息技术有限公司 | Recommendation method and device of media content, computer equipment and storage medium |
-
2021
- 2021-11-17 CN CN202111361991.2A patent/CN114201642A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108614856A (en) * | 2018-03-21 | 2018-10-02 | 北京奇艺世纪科技有限公司 | A kind of video sequence calibration method and device |
US20190297365A1 (en) * | 2018-03-23 | 2019-09-26 | Rovi Guides, Inc. | Systems and methods for increasing a likelihood of media asset consumption using recommendation chains |
CN108804492A (en) * | 2018-03-27 | 2018-11-13 | 优视科技新加坡有限公司 | The method and device recommended for multimedia object |
US20200021894A1 (en) * | 2018-07-12 | 2020-01-16 | Rovi Guides, Inc. | Systems and methods for gamification of real-time instructional commentating |
CN110781321A (en) * | 2019-08-28 | 2020-02-11 | 腾讯科技(深圳)有限公司 | Multimedia content recommendation method and device |
CN113360744A (en) * | 2020-03-04 | 2021-09-07 | 北京达佳互联信息技术有限公司 | Recommendation method and device of media content, computer equipment and storage medium |
CN111708948A (en) * | 2020-06-19 | 2020-09-25 | 北京达佳互联信息技术有限公司 | Content item recommendation method, device, server and computer readable storage medium |
CN112364186A (en) * | 2020-11-30 | 2021-02-12 | 腾讯科技(北京)有限公司 | Method, device and equipment for presenting media recommendation information and storage medium |
Non-Patent Citations (2)
Title |
---|
MAALEJ M.等: "Context Similarity Measure for Knowledge-Based Recommendation System", 《COOPERATIVE DESIGN, VISUALIZATION, AND ENGINEERING (CDVE 2017) 》, 26 August 2017 (2017-08-26), pages 77, XP047435476, DOI: 10.1007/978-3-319-66805-5_10 * |
时汉枫;: "电视媒体的新闻短视频融合传播策略", 中国传媒科技, no. 08, 15 August 2020 (2020-08-15), pages 48 - 50 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107862553B (en) | Advertisement real-time recommendation method and device, terminal equipment and storage medium | |
CN110781321B (en) | Multimedia content recommendation method and device | |
CN108763502B (en) | Information recommendation method and system | |
Fogués et al. | BFF: A tool for eliciting tie strength and user communities in social networking services | |
US20140114998A1 (en) | Determining demographics based on user interaction | |
CN110413867B (en) | Method and system for content recommendation | |
WO2019134274A1 (en) | Interest exploration method, storage medium, electronic device and system | |
CN110223186B (en) | User similarity determining method and information recommending method | |
US9864951B1 (en) | Randomized latent feature learning | |
CN113301442B (en) | Method, device, medium, and program product for determining live broadcast resource | |
CN110543598A (en) | information recommendation method and device and terminal | |
US10402465B1 (en) | Content authority ranking using browsing behavior | |
CN112104505B (en) | Application recommendation method, device, server and computer readable storage medium | |
CN111651666A (en) | User theme recommendation method and device, computer equipment and storage medium | |
CN110175264A (en) | Construction method, server and the computer readable storage medium of video user portrait | |
CN112269918B (en) | Information recommendation method, device, equipment and storage medium | |
CN113869931A (en) | Advertisement putting strategy determining method and device, computer equipment and storage medium | |
CN114201626B (en) | Multimedia recommendation method and device, electronic equipment and storage medium | |
CN111782919A (en) | Online document processing method and device, computer equipment and storage medium | |
CN113204699B (en) | Information recommendation method and device, electronic equipment and storage medium | |
CN113836388A (en) | Information recommendation method and device, server and storage medium | |
CN117312657A (en) | Abnormal function positioning method and device for financial application, computer equipment and medium | |
CN109299351B (en) | Content recommendation method and device, electronic equipment and computer readable medium | |
CN112000865B (en) | Hotword generation method, device, server and storage medium | |
CN113961811B (en) | Event map-based conversation recommendation method, device, equipment and medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |