CN113010733B

CN113010733B - Information recommendation method, device, electronic equipment and computer readable storage medium

Info

Publication number: CN113010733B
Application number: CN202011334910.5A
Authority: CN
Inventors: 刘若愚
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2020-11-25
Filing date: 2020-11-25
Publication date: 2024-05-17
Anticipated expiration: 2040-11-25
Also published as: CN113010733A

Abstract

The application provides an information recommendation method, an information recommendation device, electronic equipment and a computer readable storage medium; the method comprises the following steps: displaying a social dynamic page of the first account, wherein the social dynamic page comprises social dynamic information of interest of the first account; responding to triggering operation aiming at a dynamic information entry in a social dynamic page, determining historical dynamic information matched with the social scene where the first account is located from a plurality of historical dynamic information published in a target time period, and acquiring a video synthesized based on the matched historical dynamic information; displaying the video and displaying an interactive portal for the video; wherein the interactive portal comprises at least one of: a sharing portal, a forwarding portal, and a downloading portal. The video recommendation method and the video recommendation device can intelligently form videos with recommendation values so as to improve recommendation efficiency.

Description

Information recommendation method, device, electronic equipment and computer readable storage medium

Technical Field

The present application relates to multimedia technologies, and in particular, to an information recommendation method, an apparatus, an electronic device, and a computer readable storage medium.

Background

Artificial intelligence (AI, artificial Intelligence) is the theory, method and technique and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and expand human intelligence, sense the environment, acquire knowledge and use the knowledge to obtain optimal results.

Video recommendation is an important application of artificial intelligence, typically by providing various video editing tools, assisting users in selecting various materials to make short videos, and making recommendations according to specific application scenarios. This requires the user to perform some corresponding video editing operations, including searching for material, synthesizing material, and the like. The method has the advantages that higher video editing skill requirements are provided for users, a large amount of resources of terminal equipment of the users are consumed, the terminal equipment is heated, the time consumption is long, and the video sharing efficiency is affected.

Disclosure of Invention

The embodiment of the application provides an information recommending method, an information recommending device, electronic equipment and a computer readable storage medium, which can intelligently form video with recommending value so as to improve recommending efficiency.

The technical scheme of the embodiment of the application is realized as follows:

The embodiment of the application provides an information recommendation method, which comprises the following steps:

Displaying a social dynamic page of a first account, wherein the social dynamic page comprises social dynamic information of interest of the first account;

responding to triggering operation aiming at a dynamic information entry in the social dynamic page, determining historical dynamic information matched with the social scene where the first account is located from a plurality of historical dynamic information published in a target time period, and acquiring a video synthesized based on the matched historical dynamic information;

Displaying the video and displaying an interactive portal for the video;

wherein the interactive portal comprises at least one of: a sharing portal, a forwarding portal, and a downloading portal.

The embodiment of the application provides an information recommendation device, which comprises:

The display module is used for displaying a social dynamic page of a first account, wherein the social dynamic page comprises social dynamic information of interest of the first account; the system is also used for displaying the video and displaying an interaction inlet aiming at the video; wherein the interactive portal comprises at least one of: a sharing inlet, a forwarding inlet and a downloading inlet;

The acquisition module is used for responding to the triggering operation of the dynamic information entry in the social dynamic page, determining the historical dynamic information matched with the social scene where the first account is located from a plurality of historical dynamic information published in a target time period, and acquiring the video synthesized based on the matched historical dynamic information.

In the above solution, when the interaction portal is the sharing portal, the display module is further configured to:

In the process of displaying the video, synchronously displaying an interactive inlet of a second account corresponding to the video aiming at the second account corresponding to the face appearing in the video, and stopping displaying the interactive inlet when no triggering operation is received within waiting time; or alternatively

And displaying an interaction inlet corresponding to a second account corresponding to the face included in the video when the video is played.

In the above scheme, the display module is further configured to:

responding to the triggering operation for the forwarding entry, and displaying a forwarding editing area;

receiving forwarding content input in the forwarding editing area;

and responding to the triggering operation of the forwarding button in the forwarding editing area, and displaying the dynamic information comprising the forwarding content and the video in a dynamic information display area corresponding to the first account.

In the above scheme, the display module is further configured to:

responding to a triggering operation for the sharing portal, and displaying a sharing editing area;

Receiving the sharing content input in the sharing editing area;

and responding to the triggering operation of the sharing button in the sharing editing area, and sending dynamic information comprising the sharing content and the video to a third account associated with the sharing portal.

In the above scheme, the display module is further configured to:

displaying a time period selection area;

In response to a selection operation in the period selection area, at least one history period that is selected is taken as the target period.

In the above scheme, the obtaining module is further configured to:

Acquiring a feature vector of a social scene where the first account is located;

Invoking a neural network model based on the feature vector of the social scene and the feature vectors of the plurality of historical dynamic information to determine the adaptation probability of each historical dynamic information and the social scene, and taking the historical dynamic information exceeding an adaptation probability threshold as the historical dynamic information adapted to the social scene;

The neural network model is obtained by training based on a sample social scene and sample historical dynamic information of the sample social scene, and the interaction index of the sample historical dynamic information in the sample social scene exceeds an interaction index threshold;

wherein the interactive index comprises at least one of the following: number of praise, number of share and number of comment.

In the above scheme, the obtaining module is further configured to:

Performing descending order sorting based on interactive indexes on a plurality of historical dynamic information published by the first account in the target time period;

extracting a part of dynamic information from the descending order sequencing result to serve as historical dynamic information matched with the social scene;

In the above scheme, the display module is further configured to:

Acquiring an index of a video synthesized based on historical dynamic information of a first account, wherein the index comprises a video cover and a video link;

displaying the video cover;

And responding to the triggering operation for the video cover, acquiring the video based on the video link, and playing the video.

In the above scheme, the display module is further configured to:

sending an index request to a server so that the server generates a video link carrying an anti-theft chain based on the index request;

and receiving an index sent by the server, wherein the index comprises the video link carrying the anti-theft chain.

In the above scheme, the display module is further configured to:

sending an index request to a server to enable the server to execute the following operations:

determining the type of recommended information with the highest priority;

When the recommended information type with the highest priority is a video type synthesized based on the historical dynamic information, searching a video synthesized in advance based on the historical dynamic information of the first account;

Returning to the index when the searched pushing time in the index corresponding to the video is consistent with the current time carried by the index request;

And receiving the index returned by the server.

In the above scheme, the video and the corresponding index are synthesized by the server based on the historical dynamic information of the first account;

wherein the server synthesizes the video and corresponding index by performing the following operations:

acquiring historical dynamic information published by the first account in the target time period;

and generating an index corresponding to the video and the video based on the historical dynamic information, wherein the pushing time of the index is matched with the publishing time of the historical dynamic information.

In the above scheme, the function of synthesizing the video in the dynamic information entry is dedicated to opening the usage right to the active account, and the first account belongs to the active account;

Wherein the active account number is determined by the server by performing at least one of:

Determining a plurality of accounts which trigger a dynamic information inlet in a preset time period as candidate accounts, and screening out accounts which have published historical dynamic information in the target time period from the candidate accounts;

And selecting user accounts with preset proportions from the screened accounts as the active accounts.

An embodiment of the present application provides an electronic device, including:

a memory for storing executable instructions;

and the processor is used for realizing the information recommendation method provided by the embodiment of the application when executing the executable instructions stored in the memory.

The embodiment of the application provides a computer readable storage medium which stores executable instructions for realizing the information recommendation method provided by the embodiment of the application when being executed by a processor.

The embodiment of the application has the following beneficial effects:

And acquiring videos synthesized based on the historical dynamic information published by the user accounts, so that the videos synthesized for each user account can accurately accord with the preference of the corresponding user. Meanwhile, the diversity of videos is guaranteed due to the difference between the historical dynamic information of different users, and manual intervention is avoided, so that videos with recommendation values are automatically and intelligently formed, and recommendation efficiency is improved.

Drawings

FIG. 1A is a schematic diagram of an information recommendation system 10 according to an embodiment of the present application;

FIG. 1B is a schematic diagram of an information recommendation system 10 according to an embodiment of the present application;

FIG. 1C is a schematic diagram of an information recommendation system 10 according to an embodiment of the present application;

FIG. 1D is a block chain architecture diagram according to an embodiment of the present application;

fig. 2 is a schematic structural diagram of a server for synthesizing video according to an embodiment of the present application;

Fig. 3 is a flow chart of an information recommendation method according to an embodiment of the present application;

FIG. 4A is a schematic diagram of a social dynamic page provided by an embodiment of the present application;

FIGS. 4B-4C are schematic diagrams of dynamic information entry pages of a social application provided by embodiments of the present application;

FIG. 5 is a schematic diagram of time selection provided by an embodiment of the present application;

FIG. 6A is a schematic diagram of a video presentation page provided by an embodiment of the present application;

FIG. 6B is a schematic diagram of a first frame of video provided by an embodiment of the present application;

Fig. 6C to 6D are schematic diagrams of video intermediate frames according to an embodiment of the present application;

FIG. 6E is a schematic diagram of a video trailer provided by an embodiment of the present application;

fig. 6F is a schematic diagram of a forwarding page provided by an embodiment of the present application;

FIG. 6G is a schematic diagram of a forwarding page provided by an embodiment of the present application;

FIG. 6H is a schematic diagram of a forwarding page provided by an embodiment of the present application;

FIG. 6I is a schematic diagram of a sharing page according to an embodiment of the present application;

FIG. 6J is a schematic diagram of a sharing page according to an embodiment of the present application;

fig. 6K is a schematic diagram of a friend selection page provided by an embodiment of the present application;

fig. 6L is a schematic diagram of a friend selection page provided by an embodiment of the present application;

fig. 6M is a schematic diagram of a friend selection page provided by an embodiment of the present application;

FIG. 6N is a schematic diagram of an intermediate frame of video provided by an embodiment of the present application;

FIG. 7 is a schematic diagram of a dynamic information display area according to an embodiment of the present application;

FIG. 8 is an interactive flow diagram of an information recommendation method according to an embodiment of the present application;

fig. 9 is a schematic structural diagram of a terminal 400 according to an embodiment of the present application.

Detailed Description

The present application will be further described in detail with reference to the accompanying drawings, for the purpose of making the objects, technical solutions and advantages of the present application more apparent, and the described embodiments should not be construed as limiting the present application, and all other embodiments obtained by those skilled in the art without making any inventive effort are within the scope of the present application.

In the following description, reference is made to "some embodiments" which describe a subset of all possible embodiments, but it is to be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments and can be combined with one another without conflict.

In the following description, the terms "first/second/third" are used merely to distinguish between similar objects and do not represent a particular ordering of the objects, it being understood that the "first/second/third" may be interchanged with a particular order or precedence where allowed, to enable embodiments of the application described herein to be implemented in other than those illustrated or described herein.

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 application belongs. The terminology used herein is for the purpose of describing embodiments of the application only and is not intended to be limiting of the application.

Before describing embodiments of the present application in further detail, the terms and terminology involved in the embodiments of the present application will be described, and the terms and terminology involved in the embodiments of the present application will be used in the following explanation.

1) Today: refers to today a few years (one year or more) ago. The video of today in that year is a video synthesized by the user who published dynamic information today a few years ago based on the dynamic information published today in that year, and the purpose of the video is to further get the user to review, guide the user to publish or download the video.

2) OpenCV: is a cross-platform computer vision and machine learning software library based on BSD license (open source) release that can run on Linux, windows, android and Mac OS operating systems. The method has the characteristics of light weight and high efficiency, is composed of a series of C functions and a small number of C++ types, provides interfaces of various programming languages, and realizes various general algorithms in the aspects of image processing and computer vision.

3) FFmpeg: is an open source computer program that can be used to record, convert digital audio, video, and convert it into streams. It provides a complete solution for recording, converting and streaming audio and video.

4) Key Value storage system (CKV, cloud Key-Value): is a memory-level, persistent, low-cost, distributed Key-Value storage system. The method has the advantages of high performance, low delay, high availability and low cost, and can easily cope with the problems of mass data access, sensitive storage cost, sensitive delay and the like.

In the related art, when recommending information to a user based on historical dynamic information published by the user, particularly when recommending video, a terminal is required to acquire the historical dynamic information from a server, synthesize the historical dynamic information into video, send the synthesized video to the server for storage, acquire corresponding video links, and then issue the video, so that the user can only see the video synthesized based on the historical dynamic information. On the one hand, the method needs to spend a large amount of terminal resources, causes terminal heating, and can cause larger time delay when the terminal displays the video because the video synthesis time is longer, so that the user experience is reduced. On the other hand, when the video is synthesized based on the terminal, the user is required to manually search the materials and edit the materials so as to synthesize the personalized video which accords with the preference of the user, and the method is time-consuming and labor-consuming, has high difficulty in editing the materials and is not friendly to the user.

In order to solve the technical problems of prolonged video synthesis and high material editing difficulty in the related art, the embodiment of the application provides an information recommendation method, an information recommendation device, electronic equipment and a computer readable storage medium, which can automatically synthesize videos based on historical dynamic information and push the videos to users in real time without consuming terminal resources.

The information recommendation method provided by the embodiment of the application can be implemented by various electronic equipment, for example, the information recommendation method can be implemented by a terminal alone or by a server and the terminal in a cooperative manner. For example, the terminal alone performs the information recommendation method described below, or the terminal and the server cooperatively perform the information recommendation method described below, for example, the server synthesizes the video and the corresponding index based on the history dynamic information in advance, when the terminal sends the index request to the server, the server generates an index including a video link carrying the anti-theft chain after determining that the current time carried by the index request is consistent with the push time in the index of the video, and returns the index to the terminal, so that the terminal can acquire the video based on the index and play the video.

The electronic device for information recommendation provided by the embodiment of the application can be various types of terminal devices or servers, wherein the servers can be independent physical servers, can be server clusters or distributed systems formed by a plurality of physical servers, and can also be cloud servers for providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDNs, basic cloud computing services such as big data and artificial intelligent platforms; the terminal may be, but is not limited to, a smart phone, a tablet computer, a notebook computer, a desktop computer, a smart speaker, a smart watch, etc. The terminal and the server may be directly or indirectly connected through wired or wireless communication, which is not limited in the embodiment of the present application.

Taking a server as an example, for example, a server cluster deployed in the cloud may be used, an artificial intelligence cloud service (AIaaS, AI as a Service) is opened to users, a AIaaS platform splits several types of common AI services and provides independent or packaged services in the cloud, and the service mode is similar to an AI theme mall, and all users can access one or more artificial intelligence services provided by using the AIaaS platform through an application programming interface.

For example, one of the artificial intelligence cloud services may be an information recommendation service, that is, a cloud server encapsulates an information recommendation program provided by the embodiment of the present application. The cloud server synthesizes the video and the corresponding index based on the historical dynamic information in advance. The terminal responds to the triggering operation aiming at the dynamic information entry, an index request is sent to a cloud server, the cloud server invokes an encapsulated information recommendation program, an index comprising a video link carrying an anti-theft chain is generated, and the index is returned to the terminal, so that the terminal acquires a video based on the index and plays the video.

The information recommendation method provided by the embodiment of the application is taken as an example for the cooperative implementation of the server and the terminal. Referring to fig. 1A, fig. 1A is a schematic diagram of an architecture of an information recommendation system 10 according to an embodiment of the present application. The terminal 400 is connected to the server 200 through the network 300, and the network 300 may be a wide area network or a local area network, or a combination of both.

In some embodiments, the server 200 synthesizes the video and corresponding index based on historical dynamic information in advance. The terminal 400 responds to the triggering operation for the dynamic information entry, sends an index request to the server 200, after determining that the current time carried by the index request is consistent with the pushing time in the index of the video, the server 200 generates an index comprising a video cover and a video link carrying an anti-theft chain, and returns the index to the terminal 400, the terminal 400 displays the video cover, and the terminal 400 responds to the triggering operation for the video cover, sends a video acquisition request to the server 200 based on the video link in the index to acquire the video and display the video and the interactive entry for the video.

In some embodiments, the embodiments of the present application may also be implemented by using a blockchain technique, and referring to fig. 1B, fig. 1B is a schematic diagram of an architecture of an information recommendation system 10 provided by the embodiments of the present application. The terminal 400 transmits the historical dynamic information to the blockchain network 500 (the nodes 510-1, 510-2, and 510-3 included in the blockchain network 500 are exemplarily shown) after each time the user publishes the historical dynamic information, or transmits the historical dynamic information to the blockchain network 500 every fixed time interval, acquires the historical dynamic information of the user from the blockchain network 500 when the server synthesizes the video, and synthesizes the video based on the acquired historical dynamic information. When the server 200 receives the index request transmitted from the terminal 400, an index including a video cover and a video link carrying an anti-theft chain is generated, and the index is returned to the terminal 400; and when receiving a video acquisition request of the terminal, sending a video to the terminal.

In some embodiments, referring to fig. 1C, fig. 1C is a schematic architecture diagram of an information recommendation system 10 according to an embodiment of the present application. The information recommendation system 10 may be used to perform data sharing between nodes, and the information recommendation system 10 may include a plurality of nodes 110 (illustratively, node 110-1, node 110-2, node 110-3, and node 110-4 are shown), where the plurality of nodes 110 may refer to respective terminals in the information recommendation system 10. Each node may post video during normal operation or post video acquisition from a server and maintain shared data within the information recommendation system 10 based on the post video. In order to ensure information intercommunication within the information recommendation system 10, there may be an information connection between each node in the information recommendation system 10, and information transmission may be performed between the nodes through the above information connection. For example, when any node in the information recommendation system 10 publishes a video or obtains a video from a server and publishes the video, other nodes in the information recommendation system 10 obtain the video according to a consensus algorithm, and store the video as data in shared data, so that the data stored on all nodes in the information recommendation system 10 are consistent.

Each node in the information recommendation system 10 has a node identification corresponding thereto, and each node in the information recommendation system 10 may store node identifications of other nodes in the information recommendation system 10 so as to broadcast the generated block to other nodes in the information recommendation system 10 according to the node identifications of the other nodes. Each node may maintain a node identifier list as shown in table 1, and store the node name and the node identifier in the node identifier list. The node identifier may be an IP (Internet Protocol, protocol interconnected between networks) address and any other information that can be used to identify the node, and the IP address is only illustrated in table 1.

Node name	Node identification
		Node 1	117.114.151.174
Node 2	117.116.189.145
		…	…
Node N	119.123.789.258

Table 1 node identification list

Each node in the information recommendation system 10 stores one and the same blockchain. The blockchain is composed of a plurality of blocks, referring to fig. 1D, fig. 1D is a schematic diagram of the blockchain according to an embodiment of the present application. The block chain consists of a plurality of blocks, the starting block comprises a block head and a block main body, the block head is stored with a video characteristic value, a version number, a time stamp and a difficulty value, and the block main body is stored with a video; the next block of the starting block takes the starting block as a father block, the next block also comprises a block head and a block main body, the video characteristic value of the current block, the block head characteristic value of the father block, the version number, the timestamp and the difficulty value are stored in the block head, and the like, so that the block data stored in each block in the block chain are associated with the block data stored in the father block, and the safety of video in the block is ensured.

In some embodiments, taking the electronic device provided in the embodiments of the present application as the terminal 400, the terminal 400 implements the information recommendation method provided in the embodiments of the present application by running a computer program, where the computer program may be a native program or a software module in an operating system; may be a local (Native) Application (APP), i.e. a program that needs to be installed in an operating system to run, such as a social client; or a browser which displays video and interactive entries in the form of web pages; or a navigation applet that can be embedded in any APP. In general, the computer programs described above may be any form of application, module or plug-in.

Referring to fig. 2, fig. 2 is a schematic structural diagram of a server for synthesizing video according to an embodiment of the present application. The server comprises a video synthesis module, a material module, a first storage module, a first message module, a second message module, an agent module, a timing task module, an expiration processing module, a second storage module, an adaptation module, a recommendation module, an action bypass module and a synchronization center module. The timing task module is used for determining a user account of the video to be synthesized. The video synthesis module is used for synthesizing videos and uploading the videos. The proxy module is used for providing a fixed number of processes to draw the user account. The material module is used for providing materials. And the recommending module is used for sending the synthesized video to the terminal. The adaptation module is used for providing a link with an anti-theft chain. The synchronization center module asynchronously processes messages through the message queue. The action bypass module is used for notifying the recommendation module to modify the index of the video after the user publishes (shares or forwards) the synthesized video so as to permanently save the video and the index. The expiration processing module is used for deleting the expired video and index. The first message module is used for storing user account numbers of videos to be synthesized. The second message module is used for storing the user account of the synthesized video. The first storage module is used for storing indexes of videos. The second storage module is used for storing the synthesized video.

The information recommendation method provided by the embodiment of the present application is described below with reference to each module in the server for synthesizing video described above. As an example, the execution subject of the information recommendation method described below may be a terminal, and in particular, may be implemented by the terminal by running the various computer programs described above; of course, it will be apparent from the following understanding that the information recommendation method provided in the embodiments of the present application may also be implemented by the terminal and the server in cooperation.

Referring to fig. 3, fig. 3 is a flowchart of an information recommendation method according to an embodiment of the present application, and will be described with reference to the steps shown in fig. 3.

In step 101, a social dynamic page of the first account is displayed, wherein the social dynamic page includes social dynamic information of interest to the first account.

In some embodiments, the social dynamics includes at least one of: the method comprises the steps of enabling social dynamic information corresponding to a first account to be published by a user, enabling social dynamic information published by an account having a social relation with the first account to be published by a social application background to actively push the social dynamic information published by the account having no social relation with the first account to the first account according to portrait data of the first account. Social relationships include friend relationships, attention relationships, comment relationships (comment on social dynamic information posted by the other party), praise relationships (praise on social dynamic information posted by the other party), forwarding relationships (forward on social dynamic information posted by the other party), and the like.

In some possible examples, the social application background obtains portrait data of the user corresponding to the first account according to the content such as browsing records and published dynamic information of the user corresponding to the first account in the social application and/or the associated social application, further determines the interest and hobbies of the user corresponding to the first account based on the portrait data, pushes the dynamic information published by the account which has no social relationship with the user corresponding to the first account and is related to the interest and hobbies to the user corresponding to the first account. For example, the social application detects that the time period for browsing the dynamic information related to flowers and birds is far longer than the time period for browsing other types of dynamic information, and accordingly judges that the user corresponding to the first account is interested in the flowers and birds, so that the dynamic information recently published by the account which frequently publishes the dynamic information related to flowers and birds is pushed to the first account.

Referring to fig. 4A, fig. 4A is a schematic diagram of a social dynamic page provided by an embodiment of the present application. Social dynamic information published by users 401 and 402 is displayed in the page shown in FIG. 4A. In some possible examples, the user 401 may be an account user that is actively pushed to the user corresponding to the first account in the background and has no social relationship with the user, if the user corresponding to the first account is not interested in the user 401 and the dynamic information published by the user, the user may click on the icon 403 in fig. 4A, an option of "don't see him" and "pay attention to him" will appear, and if "don't see him" is selected, the dynamic information published by the user 401 is not pushed any more; user 402 may be an account user that has a social relationship with the user, such as friends or objects of interest.

In step 102, in response to a trigger operation for a dynamic information portal in a social dynamic page, historical dynamic information adapted to the social scene in which the first account is located is determined from a plurality of historical dynamic information published in a target period of time.

Referring to fig. 4A to 4C, fig. 4B to 4C are schematic diagrams of a dynamic information entry page of a social application according to an embodiment of the present application. The user may trigger a dynamic information entry in the social dynamic page shown in FIG. 4A. In some possible examples, the entries in the dynamic information entry pages shown in fig. 4B-4C may also be triggered. The social dynamic page/dynamic information portal page includes multiple portals, such as photo portals, log portals, message portals, the year today portal 404, etc., from which users can access, knowing different content. For example, after clicking on the photo portal in FIG. 4B or FIG. 4C, the user may see a picture previously published through the social application; after clicking on the today's year entry 404 in fig. 4B or 4C, it can be seen that the dynamic information published in the last year or years before or even further from the current same time period is displayed in the video display area in the form of video.

In some embodiments, in response to a trigger operation for a dynamic information portal in a social dynamic page, determining historical dynamic information adapted to a social scene where a first account is located from a plurality of historical dynamic information published in a target period of time may be implemented in the following manner: the terminal responds to the triggering operation aiming at the dynamic information entry in the social dynamic page, and sends an index request to the server so that the server inquires a plurality of historical dynamic information published by the first account in a target time period, and performs descending order sorting based on the interaction index on the plurality of historical dynamic information published by the first account in the target time period; extracting a part of dynamic information in front from the descending order sequencing result to be used as historical dynamic information matched with the social scene; wherein the interaction index comprises at least one of the following: number of praise, number of share and number of comment.

In some possible examples, the interaction metrics may reflect the popularity of the dynamic information, and thus, the descending order based on the interaction metrics is also a descending order based on the popularity of the dynamic information. The partial dynamic information can be one dynamic information with highest heat, or two or three dynamic information with the highest heat ranking, etc. The number of dynamic information in the partial dynamic information can be set according to practical situations, and the embodiment of the application is not limited to this. When there are multiple dynamic information in the partial dynamic information, the multiple dynamic information may be published on the same day, or published on different days, or individual dynamic information in the multiple dynamic information may be published on the same day.

In some possible examples, the descending order based on the interaction index may be ordering based on the number of praise, ordering based on the number of comments or ordering based on the number of shares, ordering based on the total amount of praise, comments and shares, or ordering after weighted summation of the number of praise, the number of comments and the number of shares, that is, respectively calculating a weighted sum corresponding to each piece of historical dynamic information published by the first account in the target time period, ordering in descending order according to the size of the weighted sum, and taking the part of dynamic information before. For example, the weight corresponding to praise is 50%, the weight corresponding to comment is 30%, and the weight corresponding to share is 20%. If the number of points corresponding to one piece of dynamic information is 30, the number of comments is 20, and the number of shares is 10, the corresponding weighted sum is 30×50% +20×30% +10×20% =23.

In some embodiments, determining historical dynamic information adapted to the social scene in which the first account is located from a plurality of historical dynamic information published in the target time period may be implemented by the following steps 1021 to 1022. The execution subject of steps 1021 to 1022 may be a terminal, or the server may return the adapted historical dynamic information to the terminal after determining the historical dynamic information through steps 1021 to 1022.

In step 1021, a feature vector of a social scene in which the first account is located is obtained.

In some possible examples, the social application obtains user behavior data (or user behavior log) such as user click data for the first account over a recent period of time (e.g., a day, 24 hours, 3 hours, etc.). And filtering the behavior data of the user to filter out the behavior data irrelevant to the social scene, and screening to obtain the behavior data relevant to the social scene. For example, if the user frequently browses dynamic information related to the current event in a recent period of time, the screened behavior data is mostly related to the current event, and the social scene can be positioned in the current event; if the user frequently browses dynamic information related to entertainment in the last period of time, most of the screened behavior data are related to entertainment, and the social scene can be positioned in entertainment. And generating feature vectors of the social scene based on the screened behavior data.

In step 1022, invoking the neural network model based on the feature vector of the social scene and the feature vectors of the plurality of historical dynamic information to determine an adaptation probability of each historical dynamic information to the social scene, and taking the historical dynamic information exceeding the adaptation probability threshold as the historical dynamic information adapted to the social scene; the neural network model is obtained by training based on a sample social scene and sample historical dynamic information of the sample social scene, and the interaction index of the sample historical dynamic information in the sample social scene exceeds an interaction index threshold; wherein the interaction index comprises at least one of the following: number of praise, number of share and number of comment.

In some possible examples, after the feature vector of the social scene is obtained, the feature vector of a plurality of historical dynamic information published by the first account in the target time period needs to be extracted. And inputting the feature vectors of the social scene and the feature vectors of the plurality of historical dynamic information into a trained neural network model, and predicting to obtain the adaptive probability of each historical dynamic information and the social scene. The historical dynamic information exceeding the fit probability threshold may be one or more.

It can be seen that the adapting probability of each historical dynamic information and the social scene is predicted through the trained neural network model, so that the historical dynamic information of the social scene adaptation of the user corresponding to the first account can be accurately determined.

In some embodiments, the terminal displays a time period selection area in response to a trigger operation for a dynamic information entry in a social dynamic page; in response to a selection operation in the period selection area, at least one history period that is selected is taken as a target period.

Referring to fig. 5, fig. 5 is a schematic diagram of time selection provided by an embodiment of the present application. The terminal displays a period selection area 501 in response to a trigger operation for a dynamic information portal. The time (year, month, day) displayed by default in the time period selection area 501 may be the last year today corresponding to the current time, for example, the current time is 2020.10.1, and the time displayed by default is 2019.10.1. The user may modify the time displayed by default in the time slot selection area 501, and after clicking the drop-down identifier in the time slot selection area 501, the year, month and day may be selected in the drop-down frame displayed, and the year, month and day may be input by filling in the blank, or the corresponding year, month and day may be selected in the calendar displayed in the time slot selection area 501. In some possible examples, the granularity of the time selection may be "week", "hour", etc., in addition to "year", "month", "day".

After the user reselects the time, the historical time period selected by the user is displayed as the target time period. For example, the target time period is 2019.10.1-2019.10.7, or 2019.10.1, the whole day. In some possible examples, the user may select multiple historical time periods as the screening time period, e.g., two days 2019.10.1 and 2018.10.1.

In some embodiments, the server synthesizes a plurality of videos in advance based on the historical dynamic information, and therefore, the user can only select further among the time periods defined by the plurality of videos, narrowing down the range of the time periods to determine the target time period. For example, when the server synthesizes video of today that year, "month" and "day" in the time period selection area 501 are fixed, and only "year" can be reselected, for example, when it is currently 2020, it can be selected as 2019 or 2018 or 2017, or the like. When the server synthesizes video of that month today, "day" in the time period selection area 501 is fixed, and "year" and "month" can be reselected. When the server synthesizes a video in which the push time of the video and the publication time of the historical dynamic information for synthesizing the video are at the same period of different years, the "month" in the period selection area 501 is fixed, the "year" can be reselected, and the "day" does not need to be selected, which defaults to all days in the fixed month. For example, the publishing time of the historical dynamic information is 2018.1.5, and the pushing time of the video synthesized based on the historical dynamic information is 2020.1.10.

In this way, the user can determine the target period according to his own needs, thereby seeing the video of today synthesized based on the historical dynamic information in the target period.

In step 103, a video synthesized based on the adapted historical dynamic information is acquired.

In some embodiments, the terminal may obtain, from the server, materials in the historical dynamic information published in the target period of time by the first account in advance, synthesize a video based on the obtained materials, and store the video in the terminal, where when the user triggers the dynamic information entry in the social dynamic page, a video corresponding to the historical dynamic information adapted to the social scene in which the user is located is obtained.

In other embodiments, the terminal transmits an index request to the server in response to a trigger operation for the dynamic information portal to determine adapted historical dynamic information and obtain a video and a corresponding index synthesized in advance based on the adapted historical dynamic information. The server synthesizes the video in advance based on the historical dynamic information published by the first account in the target time period. Because the adaptive historical dynamic information is selected from a plurality of historical dynamic information of the target time period, the corresponding video is synthesized in advance by the server.

In some possible examples, the process of the server synthesizing the video is as follows. The server acquires historical dynamic information published by the first account in a target time period; synthesizing videos in advance based on materials in the historical dynamic information of the first account, and uploading the synthesized videos to a second storage module; and generating an index corresponding to the video, and storing the index into a first storage module. The index comprises a video cover and a video link.

In some possible examples, the server generates a video link based on a storage location of the video, generates a video cover based on a preset video template, and generates a push time based on a posting time of the historical dynamic information, thereby obtaining an index comprising the video link, the video cover, and the push time. In some possible examples, the index further includes video size, video duration, video generation time, expiration time, and the like. For example, the push time is 2020.10.8, and the expiration time is 24 hours of 2020.10.8.

In some embodiments, the material in the historical dynamic information is extracted from at least one historical dynamic information published by the first account in the target time period by the server; the material in the historical dynamic information includes at least one of: pictures, text, video, audio.

In some possible examples, the material is stored in a material module of the server. The material may be in the form of pictures, text, video, audio, text+pictures (i.e., a combination of both text and pictures), text+video, text+audio, etc. When the material is 'words + pictures', the pictures can be one or more; when the material is "text+video", the video may be one or more.

In some possible examples, when the material includes a picture, the process of the server synthesizing the video is as follows. The method comprises the steps that a server determines at least one piece of historical dynamic information published by a first account in a target time period, wherein each piece of historical dynamic information comprises a picture; determining the dynamic information with highest heat in at least one piece of historical dynamic information, and acquiring a picture in the dynamic information with highest heat; generating a video intermediate frame based on the picture, and generating a video first frame and a video tail frame based on the first account; and synthesizing the first frame, the middle frame and the tail frame of the video into the video, and adding background music to the video. The first video frame and the last video frame may include user head portraits corresponding to the first account.

In some possible examples, the function of the composite video in the dynamic information portal is dedicated to opening usage rights to the active account, and the first account belongs to the active account. That is, the user corresponding to the general account number can see the dynamic information portal in each dynamic information portal page, such as the today's portal 404 in the social dynamic page shown in fig. 4A, or the today's portal 404 in the dynamic information portal pages shown in fig. 4B to 4C. But after the user corresponding to the general account number enters from the today's portal 404, only the historical dynamic information can be seen, and the video synthesized by the historical dynamic information cannot be seen as the user corresponding to the active account number. In some possible examples, the functionality of the composite video in the dynamic information portal may also open usage rights to all user accounts.

In some possible examples, the active account number is determined by the server by performing at least one of: determining a plurality of accounts which trigger a dynamic information inlet in a preset time period as candidate accounts, and screening out the accounts which have published historical dynamic information in a target time period from the candidate accounts; and selecting user accounts with preset proportions from the screened accounts as active accounts. Therefore, because only videos are needed to be synthesized for the active account, the video amount needed to be synthesized is greatly reduced, and the pressure of the server is relieved.

The preset time period refers to a period of time before the current time, and may be set in advance, for example, one week before the current time or one month before the current time. If the current time is 2020.10.8, the preset time period may be 2020.10.1-2020.10.7. Because the candidate account triggers the dynamic information entry in the preset time period, the habit of browsing the dynamic information of the candidate account can be determined according to the dynamic information entry, and therefore the active account is selected from the part of user accounts. Because the video is synthesized based on the historical dynamic information published in the target time period by the user account, the account with the historical dynamic information published in the target time period needs to be further screened from the candidate accounts. In some possible examples, the further screened account may be taken as the active account. In consideration of the fact that the number of user accounts is too large at this time, too many corresponding videos to be synthesized bring too much pressure to the server, therefore, part of user accounts are selected from the screened accounts to serve as the last active accounts, the active accounts can be randomly selected according to a preset proportion (such as 20%), and the active accounts can also be selected according to user grades or user active days.

In some possible examples, the plurality of user accounts triggered by the dynamic information entry may be further sorted in descending order according to the triggering times, and the previous user account is selected as the candidate account.

Therefore, the embodiment of the application can synthesize the video based on the historical dynamic information of the first account in advance through the server, so that the video can be displayed to the user in real time when the user triggers the dynamic information entry, and terminal resources are not consumed.

In step 104, the video is displayed.

In some embodiments, the terminal obtains an index of video synthesized based on historical dynamic information of the first account; displaying a video cover in a video display area; in response to a triggering operation for the video cover, a video acquisition request is sent to a server based on the video links in the index to acquire video from the server and play the video. In some possible examples, the terminal displays the video cover in the video display area after acquiring the index; video is obtained from the server based on the video links in the index, and the video is played. At this time, the video is directly played without triggering the video cover by the user. When the acquired index is plural, the video displayed in the video display area is plural.

In some possible examples, the video display area may be a separate page, or may be an area of a page for displaying video. The video as shown in fig. 6A presents the page. A video cover, namely a card 601 corresponding to 'that year today' is displayed in the video presentation page, and when a user clicks a play identifier 602 in the card 601, video play is started. Fig. 6B to 6E show a process of playing video, fig. 6B is a schematic diagram of a first frame of video provided by an embodiment of the present application, fig. 6C to 6D are schematic diagrams of an intermediate frame of video provided by an embodiment of the present application, and fig. 6E is a schematic diagram of a last frame of video provided by an embodiment of the present application.

In some possible examples, the terminal sends an index request to the server in response to the trigger operation. After receiving the index request, the server determines the video synthesized based on the adaptive historical dynamic information and the corresponding index; generating a video link carrying an anti-theft chain based on the original video link in the index; an index comprising a video link carrying a hotlink is sent to the terminal.

In some possible examples, the index of video also includes push time. And the terminal responds to the triggering operation and sends an index request to the server so as to request the server to acquire the index of the video corresponding to the push time consistent with the current time. After receiving the index request, the server determines the recommended information type with the highest priority; when the recommended information type with the highest priority is a video type synthesized based on the historical dynamic information, searching a video synthesized in advance based on the historical dynamic information of the first account; and when the pushing time in the index corresponding to the searched video is consistent with the current time carried by the index request, returning the index to the terminal.

The recommendation information generated by the social application can be video of the present day of the year, advertisement information, video of a commemorative day (such as birthday) and the like. When the priority of the video type synthesized based on the historical dynamic information is highest, searching whether the video synthesized based on the historical dynamic information of the first account exists, and if so, determining whether the pushing time of the index of the video synthesized in advance for the first account is consistent with the current time carried in the index request. If not, returning an index-free message to the terminal so as not to display any dynamic state in the video display area. And if the index is consistent, returning the index to the terminal. The index can be returned to the terminal after the original video link in the index is replaced by the video link carrying the anti-theft chain. For example, the pushing time of the index is 2020.10.1, the current time carried by the index request is 2020.10.1, the two times are consistent, and the server returns the index carrying the video link of the anti-theft chain to the terminal.

In some possible examples, the push time of the index matches the publication time of the historical dynamic information, i.e., both are in the same period or period of time. The push time and the publishing time may be the same day of the same month of different years (i.e. today of that year), for example, the publishing time is 2019.10.1, and the corresponding push time is 2020.10.1. The push time and the publishing time may be the same day of different months (i.e. that month is today), for example, the publishing time is 2020.9.1, and the corresponding push time is 2020.10.1. The push time and the publishing time can be respectively in the same period of different years, for example, the publishing time is 2019.9.1, the corresponding push time can be 2020.9.10 or 2020.9.20, and the like, and the corresponding push time is 9 months in 2020. The publishing time is in a target time period, and the target time period corresponds to a period of time with the same duration after the current time. For example, it is necessary to compose a video of today of the year corresponding to each day of the next week (for example 2020.10.1-2020.10.7), then it is necessary to acquire historical dynamic information published by the user in a target period (for example 2019.10.1-2019.10.7) corresponding to the next week in the last year or the previous year or more, and compose a video based on the historical dynamic information.

In step 105, an interactive portal for the video is displayed.

In some embodiments, the interactive portal includes at least one of: a sharing portal, a forwarding portal, and a downloading portal. As shown in fig. 6E, when the video is played full screen, the interactive entry may be displayed during the playing of the video (i.e., in the first frame of the video or in the middle frame of the video) or at the end of the playing of the video (i.e., in the last frame of the video), as shown by download 606, share 607, forward 608 in fig. 6E. When the video is played in a widget, an interactive portal may be displayed underneath the video or elsewhere. When the user clicks on download 606, the video is downloaded to the terminal.

In some embodiments, when the interactive portal is a sharing portal, in the process of displaying the video, synchronously displaying the interactive portal of the second account corresponding to the video aiming at the second account corresponding to the face appearing in the video, and stopping displaying the interactive portal when no triggering operation is received within the waiting time; or displaying an interaction inlet corresponding to the second account corresponding to the video aiming at the second account corresponding to the face included in the video when the video is played. The second account is an account of a user with a social relationship with the user corresponding to the first account.

Referring to fig. 6N, fig. 6N is a schematic diagram of a video intermediate frame according to an embodiment of the present application. When synthesizing the video based on the historical dynamic information, the server invokes a face recognition program, recognizes faces appearing in pictures or videos in the historical dynamic information, determines whether a user corresponding to the faces in the historical dynamic information has a social relationship with a user corresponding to the first account, and if so, acquires a second account of the user corresponding to the faces in the historical dynamic information. As shown in fig. 6N, in the process of displaying the video, when the face of the user corresponding to the second account appears, the interaction entry for the second account, that is, the sharing button 616 is displayed at the same time, and the duration of the sharing button may be 0.5 seconds, 1 second, 2 seconds, etc., which may be displayed for a period of time after the face of the user corresponding to the second account disappears when the video is played, so as to reserve a sufficient time for the triggering operation of the user. And stopping displaying the interactive entrance when the waiting time is exceeded and the triggering operation is not received yet. Or at the end of playing the video, a share 607 as in fig. 6E is displayed. Wherein the share 607 is for the second account, i.e., when the user clicks on the share 607, the video will be automatically analyzed for the second account.

Therefore, the embodiment of the application can automatically locate the user corresponding to the second account number appearing in the video of the present year through face recognition, thereby being convenient for the user to share the video of the present year to related friends, guiding the user to release the video and improving the video transmission rate.

In some embodiments, after displaying the interactive portal for the video, the terminal displays a forwarding editing region in response to a trigger operation for the forwarding portal; receiving forwarding content input in a forwarding editing area; and responding to the triggering operation of the forwarding button in the forwarding editing area, and displaying the dynamic information comprising the forwarding content and the video in the dynamic information display area corresponding to the first account.

As shown in fig. 6F, fig. 6F is a schematic diagram of a forwarding page according to an embodiment of the present application. After the user clicks on the forwarding 608 in FIG. 6E, a forwarding page is presented that includes a forwarding edit area 609. The user may or may not input forwarding content in forwarding edit area 609. As shown in fig. 6G, fig. 6G is a schematic diagram of a forwarding page according to an embodiment of the present application. The user inputs the transfer content "very happy day" in the transfer edit area 609 in fig. 6G, and clicks the transfer button below to complete transfer, and dynamic information including the transfer content and video, that is, dynamic information including the card 601 of the transfer content "very happy day" and "today of that year" is displayed in the dynamic information display area shown in fig. 7. The dynamic information display area may be a separate page, or may be an area for displaying dynamic information in a page. Thus, the user corresponding to the first account and the external visitor can see the dynamic information.

In some possible examples, dynamic information may be shared to other account numbers associated with the dynamic information while forwarding the dynamic information to the dynamic information display area. As shown in fig. 6H, fig. 6H is a schematic diagram of a forwarding page according to an embodiment of the present application. In the forward edit area 609, the first icon 610 and the second icon 611 represent social applications that are different from the present social application, and the user's account in the present social application has been associated with the accounts in both social applications. When the user clicks and selects the first icon 610/the second icon 611 and clicks and selects the forwarding button, not only the dynamic information is displayed in the dynamic information display area of the social application, but also the dynamic information is displayed in the dynamic information display area of the social application corresponding to the first icon 610/the second icon 611.

In some possible examples, the terminal sends a forwarding request carrying dynamic information including video and the first account number to the server in response to a trigger operation for a forwarding button in the forwarding edit area 609. The server obtains the forwarding request in the message queue in the synchronous center module through the action bypass module, and informs the recommendation module to modify the index corresponding to the video in the first storage module through the action bypass module, namely the identification of adding the video without deletion. Wherein, the dynamic information may or may not include the forwarding content. In this way, it can be ensured that the video required by the user is permanently stored, but not deleted. The server continuously polls and pulls the user account of the synthesized video in the second message module through the expiration processing module, so that the expired video (the expiration time of the expired video exceeds the expiration time of the index of the video, the server still does not receive a forwarding request for the video) is determined, and the expired video in the second storage module and the expired index in the first storage module are deleted through the expiration processing module. Therefore, videos which are not needed by the user can be deleted immediately, and the storage pressure is reduced.

In some embodiments, after displaying an interactive portal for a video, the terminal displays a sharing editing area in response to a trigger operation for the sharing portal; receiving the shared content input in the shared editing area; and responding to the triggering operation of the sharing button in the sharing editing area, and sending dynamic information comprising the sharing content and the video to a third account associated with the sharing portal.

Fig. 6I is a schematic diagram of a sharing page according to an embodiment of the present application, as shown in fig. 6I. After the user clicks on the share 607 in FIG. 6E, a share page is displayed, which includes a share edit area 612. The user may or may not input shared content in the shared edit area 612. Fig. 6J is a schematic diagram of a sharing page according to an embodiment of the present application, as shown in fig. 6J. The user enters the share content "very happy day" in the share edit area 612 in fig. 6J and clicks the share button below, thereby sending dynamic information including the share content and video to the third account in the other social applications associated with the share portal. The sharing portal may be multiple, and may be used to share dynamic information to dynamic information display areas in different social applications or account numbers of friends of the user in different social applications.

For example, when a user shares dynamic information to one or more friends in other social applications, the page is shown in fig. 6K, fig. 6K is a schematic diagram of a friend selection page provided by the embodiment of the present application, the user may select one shared friend in the page shown in fig. 6K, or click a multi-selection button 613 on the upper right side of fig. 6K, select multiple friends in the friend selection page shown in fig. 6L (at this time, the user may return to the page shown in fig. 6K by clicking a radio button 614 on the upper right corner of the page shown in fig. 6L), after selecting multiple friends, display the selected friends as shown in fig. 6M, and click a send button 615 on the upper right corner of fig. 6M, so that the dynamic information including the shared content and video may be sent to the friends in other social applications in the form of a message.

It can be seen that, in the embodiment of the present application, the terminal responds to the triggering operation for the dynamic information portal, acquires the video synthesized based on the historical dynamic information of the first account from the server, and acquires the video in real time and displays the video to the user without delay under the condition of not consuming terminal resources; and then, displaying an interactive entry aiming at the video, and guiding the user to publish (forward or share) and download the video while checking up the review of the user, so that the video downloading rate and the video sending rate are improved. Meanwhile, because the video is automatically synthesized in advance by the server, manual intervention is avoided, the video with recommendation value is automatically and intelligently formed, and the recommendation efficiency is improved. Because the videos are synthesized based on the historical dynamic information of the users, the synthesized videos can accurately accord with the preference of the corresponding users, and the diversity of the videos is ensured by the difference between the historical dynamic information of different users.

Referring to fig. 8, fig. 8 is an interaction flow diagram of an information recommendation method according to an embodiment of the present application. The procedure of cooperatively implementing the information recommendation method described above by the terminal and the server will be described below with reference to steps 201 to 213 in fig. 8.

In step 201, the server synthesizes a video and a corresponding index based on historical dynamic information of the first account.

In step 202, the terminal transmits an index request to the server in response to a trigger operation for the dynamic information portal.

In step 203, after determining that the current time carried by the index request is consistent with the push time in the index, the server generates a video link carrying an anti-hotlink.

In step 204, the server sends an index to the terminal comprising the video link and the video cover.

In step 205, the terminal displays a video cover.

In step 206, the terminal transmits a video acquisition request to the server based on the video link in response to the triggering operation for the video cover.

In step 207, the server transmits the video to the terminal.

In step 208, the terminal plays the video, and displays a forwarding entry for the video at the end of the video play.

In step 209, the terminal displays a forwarding edit area in response to a trigger operation for the forwarding entry.

In step 210, the terminal receives the forwarding content input in the forwarding editing region.

In step 211, the terminal sends a forwarding request carrying a first account number and dynamic information to the server in response to a triggering operation of a forwarding button for the forwarding editing area, wherein the dynamic information includes forwarding content and video.

In step 212, the terminal displays the dynamic information in the dynamic information display area corresponding to the first account.

In step 213, the server adds an undeleted identification to the video in the index of the video.

Wherein, step 212 and step 213 are not explicitly performed in sequence.

It should be noted that the foregoing steps have been described in detail in the foregoing, and are not repeated here.

In the following, an exemplary application of the embodiment of the present application in a practical application scenario will be described.

Referring to fig. 4A to 4B, fig. 4A to 4B are schematic diagrams of a dynamic portal page of a social application according to an embodiment of the present application, where the dynamic portal page includes a plurality of portals, such as a photo portal, a log portal, a message portal, a today portal 404, and so on. The entries may be displayed in a landscape orientation, in a portrait orientation, or in a tiled orientation. Through the today's portal 404 of that year, the user can see that the dynamic information posted by the user is displayed in the video display area in the form of video at the same time of the past year as the current time. For example, the current time is 24 days 2 months 2020, then the user may see through that year today's portal 404 the dynamics published by 2 month 24 daily users in years of 2019, 24 days 2 months 2018, or more.

After the user enters the dynamic portal page of the social application, the year today's portal 404 in the dynamic portal page may be triggered by clicking on, etc., the video presentation page as shown in FIG. 6A will be displayed. The card 601 corresponding to the 'today' year is displayed in the video presentation page, and when the user clicks the playing identifier 602 in the card 601, the video embedded in the card 601 starts to play. The first frame of the video is shown in fig. 6B, the middle frame of the video is shown in fig. 6C and 6D, and the last frame of the video is shown in fig. 6E.

As shown in fig. 6B to 6E, during the playing of the video, the user avatar 603 may be always displayed in the first frame, middle frame and end frame of the video. In some possible examples, the user avatar 603 and the user nickname may also be displayed simultaneously during video playback. The picture 604 and the text 605 below in fig. 6C are dynamic information published by a 24-month 2018 user, and if the dynamic information published by the user includes multiple pictures, different pictures can be presented one by one through different video intermediate frames. Fig. 6D shows the transition effect of transitioning from a video intermediate frame containing picture 604 to the next video intermediate frame containing picture 604, i.e., the text 605 disappears, and picture 604 remains. In the video footer frame of fig. 6E, a calendar related to the posting date of the dynamic information is displayed, the calendar includes a user avatar 603 and a user nickname, and there is a download 606, a share 607, and a forward 608 under the calendar.

Because the first video frame and the last video frame are generated according to the video templates, in the same time period, the video templates used for different users are the same, so that the first video frame and the last video frame corresponding to different users are the same except for the head portraits of the users, nicknames of the users and the release dates of the users. Wherein the video templates are updated at intervals.

When the user clicks on download 606, the video will be downloaded to the terminal. When the user clicks on share 607, the video will be shared to other social applications. When the user clicks on forwarding 608, the page display is as shown in fig. 6F, and fig. 6F is a schematic diagram of a forwarding page provided by an embodiment of the present application. The forwarding edit area 609 is included in fig. 6F, and when the user inputs forwarding content (or may not input any content) in the forwarding edit area 609 and clicks a forwarding button, the forwarding content and video will be forwarded and displayed in the dynamic information display area of the social application as shown in fig. 7. In the dynamic information display area, the video published by the user is presented as dynamic information together with the dynamic information recently published by other friends. Because the user does not input the forwarding content in the forwarding edit area 609, only the card 601 of "that year today" for which the video corresponds is in the dynamic information display area, and the forwarding content is not.

The respective modules in the server for composing the video today of that year will be described below.

The server comprises a video synthesis module, a material module, a first storage module, a first message module, a second message module, an agent module, a timing task module, an expiration processing module, a second storage module, an adaptation module, a recommendation module, an action bypass module and a synchronization center module.

The timing task module traverses the user account package at regular time to determine which users compose the video today in that year, namely, determine the user account of the video to be composed, and put the user account of the video to be composed into the first message module.

The video synthesis module is mainly used for synthesizing videos and uploading the videos. The video synthesis module pulls the user account number from the first message module through the proxy module, and then pulls the material from the material module, namely pulls the address of the picture in the dynamic information published by the user today in that year, and downloads the picture based on the address of the picture. The picture is processed through an OpenCV technology, video frames are synthesized based on a video template, the video frames are synthesized into video frame by utilizing cvVideoWriter functions and CVWRITEFRAME functions, and finally the video is processed through FFmpeg, such as post-processing of adding background music and the like. And uploading the video to a second storage module, writing the index of the video into the first storage module, and writing the user account of the synthesized video into a second message module.

The proxy module provides a fixed number of processes by which the user account in the first message module is processed (pulled). Therefore, the number of processes can be reduced, the problem that new processes are frequently allocated to new user accounts due to frequent process exits is avoided, and the efficiency of pulling the user accounts is improved.

The material module provides a unified material pulling interface for the video synthesis module to pull the materials (such as video, pictures, voice, text, user nicknames, user birthdays, etc.) of the synthesized video. And supporting the material of different scenes to be pulled through command word distinction.

The recommendation module is used for sending the video synthesized based on the dynamic information of today in that year to the terminal through the recommendation gateway. The recommendation gateway decides the video category (e.g., video today that year, birthday video, etc.) recommended to the user according to the recommendation priority. When no video (such as advertisement video) with higher priority than the video of today in that year is recommended, the recommendation module pulls the index in the first storage module, and judges whether the pushing time of the video is met at the moment according to the pushing time in the index. If yes, after receiving the index request sent by the terminal, returning the index to the terminal through the recommendation gateway so as to display the video in the terminal. The index comprises video links, video covers, video durations, video sizes, video generation times, push times, expiration times and the like.

The adaptation module is used for providing a link with an anti-theft chain for the terminal. When the recommending module determines that the recommending time is met, and after the index is pulled, the adapting module replaces the video link in the pulled index with a playable link with an anti-theft chain, so that the recommending module returns the index comprising the video link with the anti-theft chain to the terminal.

The synchronous center module comprises a message queue, and the messages stored in the message queue are dynamic information published by each user. The synchronous center module asynchronously processes the messages through the message queues to prevent too many messages from occupying too many resources of the next-level service (the service provided by the action bypass module).

The action bypass module receives the information in the information queue of the synchronous center module to determine the user account number which is used for publishing the dynamic video today, and informs the recommendation module to modify the index of the video corresponding to the user account number in the first storage module, and adds an undeletable identifier for the video to permanently store the video and the corresponding index.

The expiration processing module continuously polls and pulls the user account number in the second message module, and deletes the video which is expired but not exposed (the user enters the video presentation page through the today's entry to be exposed) or published, so that the storage space is saved.

The first storage module provides Get/Set/Del operation to the outside, and mainly faces to the service storage requirements of smaller average record (the maximum record size is lower than 512 KB), larger read-write quantity, lower read-write delay and high data security.

The second storage module is used for providing file object storage service.

The technical key points in the video composition process will be described below.

1. Picture synthesis video technology

The embodiment of the application synthesizes the video through OpenCV and FFmpeg. In the process of synthesizing the video, the video frames need to be adjusted according to parameters such as material correction positions, gradual changes and the like. The CvVideoWriter and CVWRITEFRAME functions in OpenCV may be used to compose a picture into a video, where the CvVideoWriter and CVWRITEFRAME functions are used to write the picture into a video file.

2. Statistics of required data resources

The embodiment of the application synthesizes videos of all active user accounts in the next week in one week, and has high real-time requirement. The method adopts the modes of judging the week activity (judging the active account in the week), timing task running user account packages and the like to narrow the range of the active account, accurately estimates the needed resource quantity in advance and ensures the video to be synthesized in the previous week.

First, a timing task is established, and an active account number (about 700 ten thousand/week) is determined; then, pulling the active account through a timing task; and finally, importing the pulled active account number into a first message module.

In the process of synthesizing the video, the time for pulling the material is 1s (second), the time for downloading the original file of the material is 9s, the time for synthesizing the video according to the original file of the material is 10-20s, the time for adding the audio track is 1-2s, the time for uploading the video is 2s, and the time for considering other links is about 36s in total.

Assuming that the magnitude of the composite video required per day is about 100 tens of thousands/day, the composite video per day is estimated to require the amount of resources: 100 thousands×36 s/(24×3600) s=416=52×8, i.e., 52 8-core servers are required to synthesize video every day.

3. Composite video strategy

The dynamic information of the active account number needed to synthesize the video is screened through the following steps.

Step 301, screening out a first user account (about 2.8 billions) of the first week of the composite video that triggered today's entry in that year;

Step 302, screening a second user account (about 3500 ten thousand) corresponding to the next week, which has been published with dynamic information (such as text with a chart) in the present year, from the first user account;

Step 303, randomly selecting 20% of user accounts from the second user accounts as active accounts (about 700 ten thousand);

step 304, selecting, for each active account, a dynamic information synthesized video (the video synthesis amount is reduced to < =100 ten thousand/day) with the highest total amount of dynamic information synthesis video (praise, comment, and forwarding) published today in the year corresponding to the next week.

4. Universalization and reusability

The architecture of the server shown in fig. 2 can support future expansion of more video types, and therefore, the architecture and storage of the server need to be generalized, and accordingly, the first storage module uses a generic video index storage architecture, and services provided by each module of the server follow a generalized principle and are not customized for specific requirements.

The structure of the server shown in fig. 2, which is reusable, is applicable not only to the composition of video today in that year, but also to the composition of other types of video, such as birthday video, advertisement video, etc. When synthesizing videos of other service types (such as advertisement videos), only the fields, such as the video service type, service information and the like, closely related to the video type need to be changed, and the other fields are all generalized. Cards corresponding to different service types can be distinguished according to the enumerated values, and the newly added service type is supported. The recommendation module has universality, and a request protocol is fixed in the interaction process of the terminal and the server.

Because of the universalization of the structure and storage of the server, customized video composition strategies can be realized, and different contents can be dynamically recommended to the user without any change of the client in the terminal.

5. Expiration deletion strategy

Because not all synthesized videos are exposed and published, and the video that is not exposed and published is outdated (exceeds the expiration time in the index) is useless, the expired video needs to be deleted.

The expiration deletion policy may ensure that published video is not deleted but is permanently stored. When a user uses the video publishing dynamics of today in that year, a non-deleting identifier is added to the video through a recommendation module so as to permanently store the video and the corresponding index. In some possible examples, if it is determined that the user clicked on the card 601 of "today's of the year" in fig. 6A, and shared the video of today's year to other social applications, the same is added with a non-deleting identifier.

The expiration deleting strategy is used for continuously polling and pulling the user account in the second message module through the expiration processing module so as to determine whether the video corresponding to the user account meets the timeout condition (namely, the current time is greater than the expiration time in the index), and deleting the video and the corresponding index when the timeout condition is met. The expiration deletion policy may be customizable and may delete expired video according to a variety of requirements.

6. Speed limiting logic

Because the video synthesis module needs to upload the video to the second storage module after synthesizing the video of today in that year, in order to avoid the influence of the uploading peak period on the uploading of the video instantly issued by other users in the social application, a speed limiting logic is set for the uploading of the video of today in that year. Specific key value pairs are set in the CKV for the video of that year today, the keys in the specific key value pairs corresponding to the video of that year today, the values including the composition latency and the deceleration period. A specific key value pair is read before each video composition to determine a composition latency and a deceleration period. After the synthesis waiting time (e.g., 15 minutes), it is determined whether the current time belongs to a deceleration period (e.g., 9 to 10 am). If the current time does not belong to the deceleration time period, starting to upload the video of the present year; if the current time belongs to the deceleration time period, the uploading speed of the video today in that year is limited.

By applying the information recommendation method provided by the embodiment of the application in social application, the user can be further hooked up to review, the user is guided to publish and download recommended videos, and the video downloading rate and the video sending rate are improved. Taking 60 ten thousand videos synthesized daily as an example, wherein the daily average exposure is 10 ten thousand, the card exposure rate is 15.78%, the card exposure click rate is 10.98%, the video schedule rate is 7.47%, and the video downloading rate is 5.97%. More portals can be added to the dynamic portal page later, and the position of the portal in the dynamic portal page today is promoted.

An exemplary structure is described below with a terminal as an example. Referring to fig. 9, fig. 9 is a schematic structural diagram of a terminal 400 provided in an embodiment of the present application, and the terminal 400 shown in fig. 9 includes: at least one processor 410, a memory 450, at least one network interface 420, and a user interface 430. The various components in terminal 400 are coupled together by a bus system 440. It is understood that the bus system 440 is used to enable connected communication between these components. The bus system 440 includes a power bus, a control bus, and a status signal bus in addition to the data bus. But for clarity of illustration the various buses are labeled in fig. 9 as bus system 440.

The Processor 410 may be an integrated circuit chip having signal processing capabilities such as a general purpose Processor, such as a microprocessor or any conventional Processor, a digital signal Processor (DSP, digital Signal Processor), or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, or the like.

The user interface 430 includes one or more output devices 431, including one or more speakers and/or one or more visual displays, that enable presentation of the media content. The user interface 430 also includes one or more input devices 432, including user interface components that facilitate user input, such as a keyboard, mouse, microphone, touch screen display, camera, other input buttons and controls.

Memory 450 may be removable, non-removable, or a combination thereof. Exemplary hardware devices include solid state memory, hard drives, optical drives, and the like. Memory 450 optionally includes one or more storage devices physically remote from processor 410.

Memory 450 includes volatile memory or nonvolatile memory, and may also include both volatile and nonvolatile memory. The non-volatile Memory may be a Read Only Memory (ROM) and the volatile Memory may be a random access Memory (RAM, random Access Memory). The memory 450 described in embodiments of the present application is intended to comprise any suitable type of memory.

In some embodiments, memory 450 is capable of storing data to support various operations, examples of which include programs, modules and data structures, or subsets or supersets thereof, as exemplified below.

An operating system 451 including system programs, e.g., framework layer, core library layer, driver layer, etc., for handling various basic system services and performing hardware-related tasks, for implementing various basic services and handling hardware-based tasks;

Network communication module 452 for reaching other computing devices via one or more (wired or wireless) network interfaces 420, exemplary network interfaces 420 include: bluetooth, wireless compatibility authentication (WiFi), and universal serial bus (USB, universal Serial Bus), etc.;

A presentation module 453 for enabling presentation of information (e.g., a user interface for operating peripheral devices and displaying content and information) via one or more output devices 431 (e.g., a display screen, speakers, etc.) associated with the user interface 430;

An input processing module 454 for detecting one or more user inputs or interactions from one of the one or more input devices 432 and translating the detected inputs or interactions.

In some embodiments, the apparatus provided in the embodiments of the present application may be implemented in software, and fig. 9 shows an information recommendation apparatus 455 stored in a memory 450, which may be software in the form of a program, a plug-in, or the like, including a display module 4551 and an acquisition module 4552.

The functions of the respective modules will be described hereinafter.

The display module 4551 is configured to display a social dynamic page of the first account, where the social dynamic page includes social dynamic information of interest of the first account; the system is also used for displaying videos and displaying an interaction inlet for the videos; wherein the interactive portal comprises at least one of: a sharing inlet, a forwarding inlet and a downloading inlet;

The acquiring module 4552 is configured to determine, from a plurality of historical dynamic information published by the first account in a target period of time, historical dynamic information adapted to the social scene in which the first account is located in response to a trigger operation for a dynamic information entry in the social dynamic page, and acquire a video synthesized based on the adapted historical dynamic information.

In some embodiments, when the interactive portal is the sharing portal, the display module 4551 is further configured to: in the process of displaying the video, synchronously displaying an interactive entry of the second account corresponding to the video aiming at the second account corresponding to the face appearing in the video, and stopping displaying the interactive entry when no triggering operation is received within the waiting time; or displaying an interaction inlet corresponding to the second account corresponding to the video aiming at the second account corresponding to the face included in the video when the video is played.

In some embodiments, the display module 4551 is further configured to: responding to the triggering operation for the forwarding entry, and displaying a forwarding editing area; receiving forwarding content input in a forwarding editing area; and responding to the triggering operation of the forwarding button in the forwarding editing area, and displaying the dynamic information comprising the forwarding content and the video in the dynamic information display area corresponding to the first account.

In some embodiments, the display module 4551 is further configured to: responding to a triggering operation for the sharing portal, and displaying a sharing editing area; receiving the shared content input in the shared editing area; and responding to the triggering operation of the sharing button in the sharing editing area, and sending dynamic information comprising the sharing content and the video to a third account associated with the sharing portal.

In some embodiments, the display module 4551 is further configured to: displaying a time period selection area; in response to a selection operation in the period selection area, at least one history period that is selected is taken as a target period.

In some embodiments, the acquiring module 4552 is further configured to: acquiring a feature vector of a social scene where a first account is located; invoking a neural network model based on the feature vector of the social scene and the feature vectors of the plurality of historical dynamic information to determine the adaptation probability of each historical dynamic information and the social scene, and taking the historical dynamic information exceeding the adaptation probability threshold as the historical dynamic information adapted to the social scene; the neural network model is obtained by training based on a sample social scene and sample historical dynamic information of the sample social scene, and the interaction index of the sample historical dynamic information in the sample social scene exceeds an interaction index threshold; wherein the interaction index comprises at least one of the following: number of praise, number of share and number of comment.

In some embodiments, the acquiring module 4552 is further configured to: performing descending order sorting based on interactive indexes on a plurality of historical dynamic information published by the first account in a target time period; extracting a part of dynamic information in front from the descending order sequencing result to be used as historical dynamic information matched with the social scene; wherein the interaction index comprises at least one of the following: number of praise, number of share and number of comment.

In some embodiments, the display module 4551 is further configured to: acquiring an index of a video synthesized based on historical dynamic information of a first account, wherein the index comprises a video cover and a video link; displaying a video cover; and responding to the triggering operation for the video cover, acquiring the video based on the video link, and playing the video.

In some embodiments, the display module 4551 is further configured to: sending an index request to a server so that the server generates a video link carrying an anti-theft chain based on the index request; and receiving an index sent by the server, wherein the index comprises a video link carrying an anti-theft chain.

In some embodiments, the display module 4551 is further configured to: sending an index request to the server to cause the server to: determining the type of recommended information with the highest priority; when the recommended information type with the highest priority is a video type synthesized based on the historical dynamic information, searching a video synthesized in advance based on the historical dynamic information of the first account; returning to the index when the pushing time in the index corresponding to the searched video is consistent with the current time carried by the index request; and receiving the index returned by the server.

In some embodiments, the video and corresponding index are synthesized by the server based on historical dynamic information of the first account; wherein the server synthesizes the video and corresponding index by performing the following operations: acquiring historical dynamic information published by a first account in a target time period; and generating an index corresponding to the video based on the historical dynamic information, wherein the pushing time of the index is matched with the publishing time of the historical dynamic information.

In some embodiments, the function of the composite video in the dynamic information portal is dedicated to opening usage rights to the active account, and the first account belongs to the active account; wherein the active account number is determined by the server by performing at least one of: determining a plurality of accounts which trigger a dynamic information inlet in a preset time period as candidate accounts, and screening out the accounts which have published historical dynamic information in a target time period from the candidate accounts; and selecting user accounts with preset proportions from the screened accounts as active accounts.

An embodiment of the present application provides a computer-readable storage medium storing executable instructions that, when executed by a processor, cause the processor to perform an information recommendation method provided by an embodiment of the present application, for example, an information recommendation method as shown in fig. 3.

In some embodiments, the computer readable storage medium may be FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; but may be a variety of devices including one or any combination of the above memories.

In some embodiments, the executable instructions may be in the form of programs, software modules, scripts, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and they may be deployed in any form, including as stand-alone programs or as modules, components, subroutines, or other units suitable for use in a computing environment.

As an example, executable instructions may, but need not, correspond to files in a file system, may be stored as part of a file that holds other programs or data, such as in one or more scripts in a hypertext markup language (HTML, hyper Text Markup Language) document, in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files that store one or more modules, sub-programs, or portions of code).

As an example, executable instructions may be deployed to be executed on one computing device or on multiple computing devices located at one site or distributed across multiple sites and interconnected by a communication network.

In summary, in the embodiment of the present application, the terminal acquires, from the server, the video synthesized based on the historical dynamic information of the first account in response to the triggering operation for the dynamic information entry, and acquires the video in real time and displays the video to the user without delay under the condition of not consuming terminal resources. And then, displaying an interactive entry aiming at the video, and guiding the user to publish (forward or share) and download the video while checking up the review of the user, so that the video downloading rate and the video sending rate are improved. Meanwhile, because the video is automatically synthesized in advance by the server, manual intervention is avoided, the video with recommendation value is automatically and intelligently formed, and the recommendation efficiency is improved. Because the videos are synthesized based on the historical dynamic information of the users, the synthesized videos can accurately accord with the preference of the corresponding users, and the diversity of the videos is ensured by the difference between the historical dynamic information of different users.

The foregoing is merely exemplary embodiments of the present application and is not intended to limit the scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and scope of the present application are included in the protection scope of the present application.

Claims

1. An information recommendation method, comprising:

displaying a social dynamic page of a first account, wherein the first account belongs to an active account, and the social dynamic page comprises social dynamic information of interest of the first account;

responding to a triggering operation for a dynamic information entry in the social dynamic page, and displaying a time period selection area;

In response to a selection operation in the period selection area, taking the selected at least one history period as a target period;

acquiring feature vectors of a social scene where the first account is located, and extracting feature vectors of a plurality of historical dynamic information published by the first account in the target time period;

Invoking a neural network model based on the feature vector of the social scene and the feature vectors of the plurality of historical dynamic information to determine the adaptation probability of each historical dynamic information and the social scene, and taking the historical dynamic information exceeding an adaptation probability threshold as the historical dynamic information adapting to the social scene, wherein the neural network model is obtained by training based on a sample social scene and sample historical dynamic information of the sample social scene, and an interaction index of the sample historical dynamic information in the sample social scene exceeds an interaction index threshold, and the interaction index comprises at least one of the following: number of praise, number of share and number of comments;

acquiring a video synthesized based on the adapted historical dynamic information;

Displaying the video and displaying an interactive portal for the video;

2. The method of claim 1, wherein when the interactive portal is the shared portal, the displaying the interactive portal for the video comprises:

3. The method of claim 1, wherein after the displaying the interactive portal for the video, the method further comprises:

receiving forwarding content input in the forwarding editing area;

4. The method of claim 1, wherein after the displaying the interactive portal for the video, the method further comprises:

Receiving the sharing content input in the sharing editing area;

5. The method of claim 1, wherein the displaying the video comprises:

displaying the video cover;

6. The method of claim 5, wherein the obtaining an index of video synthesized based on historical dynamic information of the first account comprises:

7. The method of claim 5, wherein the obtaining an index of video synthesized based on historical dynamic information of the first account comprises:

determining the type of recommended information with the highest priority; when the recommended information type with the highest priority is a video type synthesized based on the historical dynamic information, searching a video synthesized in advance based on the historical dynamic information of the first account; returning to the index when the searched pushing time in the index corresponding to the video is consistent with the current time carried by the index request;

And receiving the index returned by the server.

8. The method of claim 5, wherein the video and corresponding index are synthesized by a server based on historical dynamic information of the first account;

9. The method of claim 1, wherein the function of synthesizing the video in the dynamic information portal is dedicated to opening usage rights to an active account;

10. An information recommendation device, characterized by comprising:

the display module is used for displaying a social dynamic page of a first account, wherein the first account belongs to an active account, and the social dynamic page comprises social dynamic information of interest of the first account;

The acquisition module is used for responding to the triggering operation of the dynamic information entry in the social dynamic page and displaying a time period selection area; in response to a selection operation in the period selection area, taking the selected at least one history period as a target period; acquiring feature vectors of a social scene where the first account is located, and extracting feature vectors of a plurality of historical dynamic information published by the first account in the target time period; invoking a neural network model based on the feature vector of the social scene and the feature vectors of the plurality of historical dynamic information to determine the adaptation probability of each historical dynamic information and the social scene, and taking the historical dynamic information exceeding an adaptation probability threshold as the historical dynamic information adapted to the social scene; the neural network model is obtained by training based on a sample social scene and sample historical dynamic information of the sample social scene, wherein an interaction index of the sample historical dynamic information in the sample social scene exceeds an interaction index threshold, and the interaction index comprises at least one of the following: number of praise, number of share and number of comments;

the acquisition module is further used for acquiring a video synthesized based on the adaptive historical dynamic information;

the display module is also used for displaying the video and displaying an interaction inlet aiming at the video; wherein the interactive portal comprises at least one of: a sharing portal, a forwarding portal, and a downloading portal.

11. An electronic device, comprising:

a memory for storing executable instructions;

A processor for implementing the information recommendation method according to any one of claims 1 to 9 when executing executable instructions stored in said memory.

12. A computer readable storage medium storing executable instructions for causing a processor to perform the information recommendation method according to any one of claims 1 to 9.