CN113411618B

CN113411618B - Data processing method and device based on social application and computer storage medium

Info

Publication number: CN113411618B
Application number: CN202011356955.2A
Authority: CN
Inventors: 陈春勇; 杨琳
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2020-11-26
Filing date: 2020-11-26
Publication date: 2024-03-22
Anticipated expiration: 2040-11-26
Also published as: CN113411618A

Abstract

The embodiment of the application discloses a data processing method, device and computer readable storage medium based on social application, wherein the method comprises the following steps: responding to the triggering operation of the question and answer adding control, displaying at least one candidate question and answer item, wherein the at least one candidate question and answer item is generated based on network media data; responding to the adding operation aiming at least one candidate question-answer item, and displaying a release control; and responding to the triggering operation for the release control, and releasing the interactive service for the added candidate question-answer items. By adopting the method and the device, the display effect of the social application can be enriched, the data interaction efficiency can be improved, and the correctness of the displayed question-answer interaction data can be ensured because candidate question-answer items can be automatically generated.

Description

Data processing method and device based on social application and computer storage medium

Technical Field

The present disclosure relates to the field of internet technologies, and in particular, to a social application-based data processing method, apparatus, device, and computer-readable storage medium.

Background

With the rapid development of mobile internet technology and various emerging technologies, social applications are also coming to the rapid development. With development, social applications are increasingly concerned about their active retention, and in order to promote their active retention, each large social application is continually added with novel interactive playing methods.

The efficient live interaction can promote active retention of social applications, and the current common live interaction mode is as follows: firstly, a main broadcasting dictation topic drives audience to discuss; secondly, the anchor manually inputs topics in the text chat interaction area, and then the audience selects option answers corresponding to the topics. Both live broadcast interaction modes require that the audience inputs option labels or own ideas in the corresponding text chat interaction areas, namely the display effect of the discussion topics is single, which results in low data interaction efficiency and easy occurrence of errors of the input topics or discussion results.

Disclosure of Invention

The embodiment of the application provides a data processing method, device and equipment based on social application and a computer readable storage medium, which not only can enrich the display effect of live interaction in the social application, but also can improve the interaction efficiency and ensure the correctness of the displayed question-answer interaction data.

In one aspect, an embodiment of the present application provides a social application-based data processing method, including:

responding to the triggering operation of the question and answer adding control, displaying at least one candidate question and answer item, wherein the at least one candidate question and answer item is generated based on network media data;

responding to the adding operation aiming at least one candidate question-answer item, and displaying a release control;

and responding to the triggering operation for the release control, and releasing the interactive service for the added candidate question-answer items.

An aspect of an embodiment of the present application provides a data processing apparatus based on a live broadcast application, including:

the first response module is used for responding to the triggering operation of the question and answer adding control, displaying at least one candidate question and answer item, and generating at least one candidate question and answer item based on network media data;

the second response module is used for responding to the adding operation aiming at least one candidate question-answer item and displaying a release control;

and the third response module is used for responding to the triggering operation for the release control and releasing the interactive service for the added candidate question-answer items.

Wherein the at least one candidate question-answer item comprises a candidate hot question-answer item;

a first response module comprising:

The first display unit is used for responding to the triggering operation of the question and answer adding control, and displaying a heat mark used for representing that the question and answer type is a heat question and answer type and candidate heat question items; the question-answer type of the candidate hot question item is a hot question-answer type;

the second display unit is used for responding to the triggering operation of the candidate hot topic items and displaying question and answer item detail areas and question and answer editing controls of the candidate hot topic items; the question and answer item detail area comprises a candidate heat question item, a candidate heat option item and a candidate answer paraphrasing item;

the activation function unit is used for responding to the triggering operation aiming at the question and answer editing control, activating the self-defining function of the detail area of the question and answer item, and updating the candidate hot question item, the candidate hot option item and the candidate answer paraphrasing item based on the self-defining function;

and the first determining unit is used for determining the updated candidate hot question item, the updated candidate hot option item and the updated candidate answer interpretation item as candidate hot question and answer items and displaying the candidate hot question and answer items.

Wherein the at least one candidate question item comprises a target candidate question item;

A second response module comprising:

the first response unit is used for responding to the triggering operation of the candidate adding control aiming at the target candidate question-answering item and switching the candidate adding control into an added control; the added control is used for representing that the target candidate question-answer item is the added candidate question-answer item;

the second response unit is used for responding to the triggering operation of the question-answer return control and displaying a list of to-be-issued question-answer items; the to-be-issued question and answer item list comprises the added candidate question and answer items and the issued controls corresponding to the added candidate question and answer items.

Wherein the added candidate question-answer items comprise candidate hotness item items and candidate hotness item items;

a third response module comprising:

the third display unit is used for responding to the triggering operation for the release control, displaying a hot question-answering sub-page and displaying the interactive service of the added candidate question-answering items through the hot question-answering sub-page;

the fourth display unit is used for displaying the hot interaction detail sub-page if the triggering operation aiming at the hot question sub-page is responded; the hot interaction detail subpage comprises candidate hot title items, candidate hot option items and interaction data corresponding to the candidate hot option items.

Wherein the first response module comprises:

the first generation unit is used for responding to the triggering operation of the question and answer adding control and generating a question and answer acquisition request;

and a fifth display unit for displaying at least one candidate question-answer item based on the heartbeat trigger period and the question-answer acquisition request.

Wherein the at least one candidate question item includes a first hot question item and a second hot question item;

a fifth display unit including:

a first display subunit, configured to display a first heat question-answer item at a first time based on the question-answer acquisition request;

the first determining subunit is used for determining a second moment according to the heartbeat trigger period and the first moment; the time interval between the first time and the second time is equal to the heartbeat trigger period, and the second time is later than the first time;

the second display subunit is used for continuously displaying the first hot question and answer item if the second hot question and answer item acquired at the second moment is the same as the first hot question and answer item;

and the third display subunit is used for switching the first hot question and answer item to the second hot question and answer item and displaying the second hot question and answer item if the second hot question and answer item is different from the first hot question and answer item.

a first response module comprising:

the second generation unit is used for responding to the triggering operation of the question and answer adding control and generating a question and answer acquisition request;

the second generation unit is also used for acquiring network media data based on the question-answer acquisition request and inputting the network media data into the hotness recommendation model;

a second determining unit, configured to determine real-time popularity media data from the network media data based on the popularity recommendation model;

and the output item unit is used for outputting candidate hotness question-answering items according to the real-time hotness media data.

The heat recommendation model comprises a base layer, a recommendation layer and a sequencing layer;

a second determination unit including:

the extraction feature subunit is used for extracting real-time features of the network media data and user features of the user history behavior data based on the base layer and inputting the real-time features and the user features into the recommendation layer; the user historical behavior data refers to user behavior data which has performed interactive operation on the network media data;

the second determining subunit is used for identifying real-time characteristics and user characteristics based on the recommendation layer, determining candidate hotness media data from the network media data according to the identification result, and inputting the candidate hotness media data into the ordering layer;

And the third determining subunit is used for carrying out sorting processing on the candidate hot media data based on the sorting layer, and determining real-time hot media data from the candidate hot media data according to the sorting result.

Wherein extracting the feature sub-unit comprises:

the first processing subunit is used for carrying out semantic identification processing on the network media data based on the semantic identification component in the base layer to obtain real-time characteristics of the network media data;

and the second processing subunit is used for carrying out behavior recognition processing on the user historical behavior data based on the user recognition component in the base layer to obtain the user characteristics of the user historical behavior data.

The recommendation layer comprises a user recommendation layer, an article recommendation layer and a content recommendation layer; the candidate popularity media data comprises user popularity media data, article popularity media data and content popularity media data;

a second determination subunit comprising:

the first identification subunit is used for identifying user characteristics based on the user recommendation layer, carrying out user popularity scoring on the network media data according to the user identification result, and determining the user popularity media data from the network media data according to the user popularity scoring;

the second identification subunit is used for identifying the user characteristics based on the item recommendation layer, carrying out item popularity scoring on the network media data according to the item identification result, and determining the item popularity media data from the network media data according to the item popularity scoring;

The third identification subunit is used for identifying real-time characteristics based on the content recommendation layer, and carrying out content marking processing on the network media data according to the content identification result to obtain a content marking result;

and the third recognition subunit is further used for determining content hot media data from the network media data according to the user characteristics and the content labeling result.

The ordering layer comprises a data de-duplication layer and a data ordering layer;

a third determination subunit comprising:

the third processing subunit is used for carrying out de-duplication processing on the user heat media data, the article heat media data and the content heat media data based on the data de-duplication layer to obtain target candidate heat media data;

and the fourth processing subunit is used for inputting the target candidate heat media data into the data ordering layer and ordering the target candidate heat media data based on the data ordering layer.

The data sorting layer comprises a forwarding data sorting layer, a comment data sorting layer and a forward data sorting layer;

the fourth processing subunit is specifically configured to input the target candidate popularity media data into the forwarding data ordering layer, the comment data ordering layer and the forward data ordering layer respectively;

The fourth processing subunit is further specifically configured to determine a forwarding probability of the target candidate hot media data based on the forwarding data ordering layer, determine a comment probability of the target candidate hot media data based on the comment data ordering layer, and determine a forward probability of the target candidate hot media data based on the forward data ordering layer;

the fourth processing subunit is further specifically configured to perform weighted fusion on the forwarding probability, the comment probability and the forward probability to obtain a fusion probability of the target candidate heat media data;

and the fourth processing subunit is further specifically configured to sort the target candidate popularity media data based on the fusion probability.

Wherein the output item unit includes:

the heat obtaining subunit is used for obtaining heat comment data of the real-time heat media data and obtaining heat keywords of the heat comment data;

and the output item subunit is used for outputting at least one candidate question-answer item according to the hotness key words.

In one aspect, the present application provides a computer device comprising: a processor, a memory, a network interface;

the processor is connected to the memory and the network interface, where the network interface is used to provide a data communication function, the memory is used to store a computer program, and the processor is used to call the computer program to execute the method in the embodiment of the present application.

In one aspect, embodiments of the present application provide a computer readable storage medium storing a computer program, where the computer program includes program instructions, where the program instructions are executed by a processor to perform a method in an embodiment of the present application.

In one aspect, embodiments of the present application provide a computer program product or computer program comprising computer instructions stored in a computer-readable storage medium; the processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the methods in the embodiments of the present application.

In the embodiment of the application, at least one candidate question-answer item can be displayed by responding to the triggering operation of the question-answer adding control, wherein the at least one candidate question-answer item is a question-answer item generated based on network media data; further, in response to an add operation for at least one candidate question-answer item, a release control may be displayed; further, in response to a triggering operation for the publishing control, an interactive service for the added candidate question and answer item can be published in the social application. According to the method and the device, the display effect of the social application can be enriched, the data interaction efficiency can be improved, and the accuracy of the displayed question-answer interaction data can be guaranteed because candidate question-answer items can be automatically generated.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic diagram of a system architecture according to an embodiment of the present application;

FIG. 2 is a schematic view of a scenario of social application-based data processing provided in an embodiment of the present application;

FIG. 3 is a schematic view of a scenario of social application-based data processing provided in an embodiment of the present application;

FIG. 4 is a schematic flow chart of a data processing method based on a social application according to an embodiment of the present application;

FIG. 5 is a schematic view of a scenario of social application-based data processing provided by an embodiment of the present application;

FIG. 6 is a schematic view of a scenario of social application-based data processing provided by an embodiment of the present application;

FIG. 7 is a schematic flow chart of a data processing method based on a social application according to an embodiment of the present application;

FIG. 8 is a schematic view of a scenario of social application-based data processing provided by an embodiment of the present application;

FIG. 9 is a schematic flow chart of a data processing method based on a social application according to an embodiment of the present application;

FIG. 10 is a schematic structural diagram of a heat recommendation model according to an embodiment of the present disclosure;

FIG. 11 is a schematic diagram of a model structure of a sort framework provided in an embodiment of the present application;

FIG. 12 is a schematic diagram of a data processing apparatus based on a social application according to an embodiment of the present application;

fig. 13 is a schematic structural diagram of a computer device according to an embodiment of the present application.

Detailed Description

The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, are intended to be within the scope of the present application.

For ease of understanding, the following simple explanation of partial nouns is first made:

artificial intelligence (Artificial Intelligence, AI) is the theory, method, technique and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and extend human intelligence, sense the environment, acquire knowledge and use the knowledge to obtain optimal results. In other words, artificial intelligence is an integrated technology of computer science that attempts to understand the essence of intelligence and to produce a new intelligent machine that can react in a similar way to human intelligence. Artificial intelligence, i.e. research on design principles and implementation methods of various intelligent machines, enables the machines to have functions of sensing, reasoning and decision.

The artificial intelligence technology is a comprehensive subject, and relates to the technology with wide fields, namely the technology with a hardware level and the technology with a software level. Artificial intelligence infrastructure technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation/interaction systems, mechatronics, and the like. The artificial intelligence software technology mainly comprises a computer vision technology, a voice processing technology, a natural language processing technology, machine learning/deep learning and other directions.

Natural language processing (Nature Language processing, NLP) is an important direction in the fields of computer science and artificial intelligence. It is studying various theories and methods that enable effective communication between a person and a computer in natural language. Natural language processing is a science that integrates linguistics, computer science, and mathematics. Thus, the research in this field will involve natural language, i.e. language that people use daily, so it has a close relationship with the research in linguistics. Natural language processing techniques typically include text processing, semantic understanding, machine translation, robotic questions and answers, knowledge graph techniques, and the like.

Machine Learning (ML) is a multi-domain interdisciplinary, involving multiple disciplines such as probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory, etc. It is specially studied how a computer simulates or implements learning behavior of a human to acquire new knowledge or skills, and reorganizes existing knowledge structures to continuously improve own performance. Machine learning is the core of artificial intelligence, a fundamental approach to letting computers have intelligence, which is applied throughout various areas of artificial intelligence. Machine learning and deep learning typically include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, induction learning, and the like.

The scheme provided by the embodiment of the application relates to artificial intelligence natural language processing technology, machine learning technology and other technologies, and is specifically described through the following embodiments.

Referring to fig. 1, fig. 1 is a schematic diagram of a system architecture according to an embodiment of the present application. As shown in fig. 1, the system may include an application server 10a, a social terminal 10b, social terminals 10c, …, a social terminal 10d, and a user terminal cluster, which may include: user terminal cluster 100b connected to social terminal 10b, user terminal clusters 100c, … connected to social terminal 10c, and user terminal cluster 100d connected to social terminal 10 d. It will be appreciated that the social terminals described above may include one or more social terminals, the number of which will not be limited herein; the above-mentioned user terminal clusters may comprise one or more user terminal clusters, the number of which will not be limited here.

The user terminal cluster 100b may include the user terminal 101b, the user terminals 102b, …, and the user terminal 103b, the user terminal cluster 100c may include the user terminal 101c, the user terminals 102c, …, and the user terminal 103c, and the user terminal cluster 100d may include the user terminal 101d, the user terminals 102d, …, and the user terminal 103d. It will be appreciated that the user terminal cluster 100b may comprise one or more user terminals, the user terminal cluster 100c may comprise one or more user terminals, and the user terminal cluster 100d may comprise one or more user terminals, the number of which will not be limited here.

Among these, a communication connection may exist between the user terminal clusters, for example, a communication connection exists between the user terminal 101b and the user terminal 102b, a communication connection exists between the user terminal 101b and the user terminal 102c, and a communication connection exists between the user terminal 101b and the user terminal 103 c. Any user terminal in the user terminal cluster may have a communication connection with the social terminal, for example, a communication connection between the user terminal 101b and the social terminal 10b, a communication connection between the user terminal 101b and the social terminal 10c, and a communication connection between the user terminal 101b and the social terminal 10 d. And there may also be a communication connection between the social terminals, for example, a communication connection between social terminal 10b and social terminal 10c, and a communication connection between social terminal 10b and social terminal 10 d.

Any user terminal in the user terminal cluster may have a communication connection with the application server 10a, for example, a communication connection between the user terminal 101b and the application server 10a, a communication connection between the user terminal 101c and the application server 10a, and a communication connection between the user terminal 101d and the application server 10 a. Meanwhile, any social terminal may have a communication connection with the application server 10a, for example, a communication connection between the social terminal 10b and the application server 10a, a communication connection between the social terminal 10c and the application server 10a, and a communication connection between the social terminal 10d and the application server 10 a.

It should be understood that the communication connection is not limited to a connection manner, and may be directly or indirectly connected through a wired communication manner, may be directly or indirectly connected through a wireless communication manner, and may also be connected through other connection manners, which is not limited herein.

It should be appreciated that the application server 10a shown in fig. 1 may be in the background of the social application Y, and the social terminal 10b, the social terminals 10c, …, the social terminal 10d, and the user terminal cluster are all installed with the social application Y. When a terminal (including any of the user terminals or social terminals described above) runs a social application Y, data interaction with the application server 10a may be performed. The social application Y may be an independent application, or may be an embedded sub-application integrated in a certain client (e.g., a live client, an instant messaging client, a game client, a social client, an educational client, a multimedia client, etc.), which is not limited herein.

Taking a live interaction scene in a live application as an example, the following description will refer to the process described below for other application scenes. Social terminal 10B in fig. 1 may be a terminal corresponding to anchor B, social terminal 10C may be a terminal corresponding to anchor C, and social terminal 10D may be a terminal corresponding to anchor D. The user terminal 101B, the user terminals 102B, …, and the user terminal 103B may be terminals corresponding to live audience of the anchor B, the user terminal 101C, the user terminals 102C, …, and the user terminal 103C may be terminals corresponding to live audience of the anchor C, …, and the user terminal 101D, the user terminals 102D, …, and the user terminal 103D may be terminals corresponding to live audience of the anchor D, respectively.

For ease of description and understanding, the anchor B, social terminal 10B, and user terminal 101B are described as examples. When the anchor B wants to add a live question and answer before or at the time of the start of the play, the social terminal 10B may be operated so that the social terminal 10B may transmit a question and answer acquisition request to the application server 10a through the live application. After the application server 10a obtains the question and answer obtaining request, network media data can be obtained according to the question and answer obtaining request, and then the network media data is input into a heat recommendation model trained in advance; performing feature extraction and feature recognition on the network media data based on the pre-trained popularity recommendation model, and determining candidate popularity media data from the network media data according to recognition results; then the application server 10a performs sorting processing on the candidate hot media data, and determines real-time hot media data from the candidate hot media data according to the sorting result; finally, the application server 10a outputs at least one candidate question and answer item according to the real-time hot media data and the big data of the whole network.

Subsequently, the application server 10a may send the generated at least one candidate question-answer item to the social terminal 10b, and at the same time, may store the acquired network media data, the feature corresponding to the network media data, and the at least one candidate question-answer item in association with the database. When the question and answer acquisition request is acquired again and the network media data acquired according to the question and answer acquisition request substantially coincides with the network media data stored in the database, the application server 10a may return at least one candidate question and answer item to the social terminal (may be the social terminal 10b, the social terminal 10c or the social terminal 10 d) that sent the question and answer acquisition request. The database may be regarded as an electronic filing cabinet, where electronic files (in this application, the electronic files may refer to network media data, features corresponding to the network media data, and at least one candidate question-answer item) are stored, and the application server 10a may perform operations such as adding, querying, updating, and deleting the network media data, the features corresponding to the network media data, and the at least one candidate question-answer item in the file. A "database" is a collection of data stored together in a manner that can be shared with multiple users, with as little redundancy as possible, independent of the application.

When the social terminal 10b receives at least one candidate question item transmitted from the application server 10a, the at least one candidate question item may be displayed on a screen, and the at least one candidate question item may include a candidate hotness item, and a candidate answer paraphrase item corresponding to the candidate hotness item. Subsequently, the anchor B may select a hot question-answer item to be published from among the at least one candidate question-answer item, or may edit the at least one candidate question-answer item, and then publish the edited question-answer item. After the anchor B issues the hot question and answer item or the edited question and answer item, a hot question and answer page may be displayed on a screen corresponding to the social terminal 10B, and a hot question and answer page may also be displayed on a screen corresponding to the user terminal 101B who is watching the anchor B, and the viewer user corresponding to the user terminal 101B may determine his or her own option based on the hot question and answer page to join the question and answer discussion of the anchor B. The anchor B can determine the number of audiences participating in the questions and the option conditions of the audiences based on the hot question-answering sub-page, so that the anchor B can interact with the audiences better.

Alternatively, if the trained popularity recommendation model is stored locally at the social terminal 10b, the process of determining at least one candidate question-answer item based on the network media data may be implemented locally. Because training the popularity recommendation model involves a large number of offline calculations, the popularity recommendation model local to the social terminal 10b may be sent to the social terminal 10b after the completion of training by the application server 10 a.

It may be appreciated that the method provided in the embodiments of the present application may be performed by a computer device, and the application server 10a and the social terminals (including the social terminal 10b, the social terminal 10c and the social terminal 10 d) in the embodiments of the present application may be all computer devices. Computer devices include, but are not limited to, social terminals or application servers. The application server may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing 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 intelligence platforms. The social 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 television, a smart watch, etc. The number of social terminals and servers is not limited, and the social terminals and servers can be directly or indirectly connected through wired or wireless communication, which is not limited herein.

Further, referring to fig. 2, fig. 2 is a schematic view of a scenario of data processing based on a social application according to an embodiment of the present application. As shown in fig. 2, at 9-point 11 minutes 25 seconds, the social terminal 20a is running a live application, and the live page 20b may display a host 20c, a number of viewers 20d (e.g., popularity 20013 shown in fig. 2), a live topic 20e (e.g., "several interesting topics this discussion" shown in fig. 2), and a question and answer addition control 20f. It will be appreciated that the scenario illustrated in fig. 2 is a scenario in which the anchor 20c adds questions and answers after the anchor 20c is opened, and in practical application, the anchor 20c may add questions and answers before the anchor, and the specific implementation process may be described below.

When the anchor 20c clicks the question and answer adding control 20f, the social terminal 20a may display a question and answer adding page 20g in response to a trigger operation for the question and answer adding control 20f, as shown in fig. 2. The question and answer adding page 20g set by the live application system is a blank page, so before the anchor 20c has not added a question and answer, the question and answer adding page 20g may display a prompt 20h, such as "the live broadcast has not added a question and answer" shown in fig. 2, and at the same time, the question and answer adding page 20g may display a gray-scale added question and answer 20z (gray may indicate that the current question and answer adding page has not added a question and answer) and a question and answer adding control 20i. When the anchor 20c clicks the question and answer adding control 20i, the social terminal 20a may display a question and answer adding page 20j, in which at least one candidate question and answer item, a hotness flag for marking a question and answer type as a hotness question and answer type, and a candidate adding control may be included.

It is to be understood that the question and answer adding page 20g is a sub-page or floating window page (which is a sub-page independently displayed on the live page 20 b) of the live page 20b, and the question and answer adding page mentioned below is a sub-page independently displayed on the live page 20b, like the question and answer adding page 20 g.

The at least one candidate question-answering item may include item 20k shown in fig. 2, i.e., item "do empty stomach can drink milk? "and item 20m, item"? "; the heat mark as described above is a heat mark 20n and a heat mark 20p shown in fig. 2; the candidate adding controls described above are, for example, the candidate adding control 20q and the candidate adding control 20r in fig. 2. Wherein the heat flag 20n may characterize the item "do empty can drink milk? "is a hot question type, the candidate add control 20q may characterize the item" do empty can drink milk? "is a candidate item state. Is the heat signature 20p characterisable for the item "Piao" worth the bean cotyledon 9 minutes? "is a hot question-answer type," is a candidate addition control 20r can characterize the item "Piao" worth the bean 9 minutes? "is a candidate item state.

It should be noted that, the at least one candidate question-answer item is a question-answer item generated based on the media data of the whole network (i.e. the network media data), and is not a question-answer item preset in the live broadcast application; the network media data may be news or events, articles, videos, public numbers, etc. that are all over the network.

If the anchor 20c clicks the candidate adding control 20q, the social terminal 20a responds to the adding operation for the item 20k, that is, responds to the triggering operation of the candidate adding control 20q for the item 20k, please refer to fig. 3 together, and fig. 3 is a schematic view of a data processing scenario based on the social application provided in the embodiment of the present application. As shown in fig. 3, the social terminal 20a switches the candidate adding control 20q in fig. 2 to an added control 30a, and the added control 30a may characterize the item "do empty stomach can drink milk? The item status of "is the item status to be issued.

The anchor 20c clicks the question and answer return control 30b in the question and answer addition page 30c, and the social terminal 20a may display a question and answer addition page 30d as shown in fig. 3, and the question and answer addition page 30d may include a normally displayed added question and answer 30e (the normally displayed added question and answer 30e may characterize that the question and answer addition page 30d has been added with a question and answer, as in the item 20k illustrated in fig. 3), an addition control 20i, a heat mark 20n, an item 20k, and a release control 30f. When the anchor 20c clicks the posting control 30f, the social terminal 20a may post in the live application "do it empty to drink milk" for the item in response to a triggering operation for the posting control 30 f? "the interactive service, the specific implementation process is described below.

Further, referring to fig. 4, fig. 4 is a flow chart of a data processing method based on social applications according to an embodiment of the present application. The data processing method may be executed by a computer device, which may be the social terminal or the application server 10a shown in fig. 1, so that the data processing method may also be executed by the social terminal and the application server 10a together, and in the embodiment of the present application, the method is described as being executed by the social terminal. As shown in fig. 4, the data processing procedure includes the following steps.

Step S101, in response to a triggering operation for the question and answer adding control, displaying at least one candidate question and answer item, the at least one candidate question and answer item being generated based on the network media data.

Specifically, the at least one candidate question-answer item includes a candidate hot question-answer item; responding to a triggering operation aiming at the question and answer adding control, and displaying a heat mark used for representing that the question and answer type is a heat question and answer type and candidate heat question items; the question-answer type of the candidate hot question item is a hot question-answer type; responding to triggering operation for the candidate hot topic items, and displaying question and answer item detail areas and question and answer editing controls for the candidate hot topic items; the question and answer item detail area comprises a candidate heat question item, a candidate heat option item and a candidate answer paraphrasing item; responding to the triggering operation aiming at the question and answer editing control, activating the self-defining function of the question and answer item detail area, and updating the candidate hot question item, the candidate hot option item and the candidate answer paraphrasing item based on the self-defining function; and determining the updated candidate hot question item, the updated candidate hot option item and the updated candidate answer paraphrase item as candidate hot question and answer items, and displaying the candidate hot question and answer items.

In response to the triggering operation of the question and answer adding control, a hotness identifier for characterizing that the question and answer type is a hotness question and a specific reality process of the candidate hotness question item are displayed, please refer to the description in the embodiment corresponding to fig. 2, and the detailed description is omitted here.

It is to be understood that the above-mentioned "item" (including the candidate hot question item, the candidate hot option item, and the candidate answer paraphrasing item) may be one of text, image, or icon, or may be a combination of text and image, or a combination of text and icon, or the like, which is not limited to the type of item, and may be set according to the actual application scenario. Likewise, the items described below are also the same. In the embodiment of the present application, description will be made taking an item as an example.

Referring back to fig. 2, the question and answer add page 20j may include candidate hot question and answer items, hot identification for marking the question and answer type as hot question and candidate add controls, and may also include candidate question and answer items and candidate add controls for the candidate question and answer items, where the candidate question and answer items may include the item "do this generation young stress big? "and the item" which is you expected in national celebration grade hospital line? ". The candidate add control 20s may characterize the term "is this generation young stressed big? "the item status is the candidate item status, the candidate adding control 20t can characterize the item" what is you expecting in national celebration court line big? "is a candidate item state.

In the embodiment of the application, the candidate hot question and answer items may include a candidate hot question item, a candidate hot option item and a candidate answer paraphrasing item, and similarly, the candidate question and answer items may include a candidate question item, a candidate option item and a candidate answer paraphrasing item. It will be appreciated that the question and answer adding page 20j displays only candidate question items and candidate question items, but the candidate question items and candidate answer paraphrasing items are generated for the candidate question items, and similarly, the candidate item and candidate answer paraphrasing items are generated for the candidate question items, and further, when the candidate question items are triggered as will be described below, the social terminal 20a displays a question and answer item detail area for the candidate question items, the question and answer item detail area including the candidate question items, the candidate question item and the candidate answer paraphrasing items, and is described herein as the candidate question and answer items.

Referring back to fig. 2, the question and answer add page 20j may also include a question and answer selection control (such as the question and answer selection control 20v and the question and answer selection control 20w shown in fig. 2) and a question and answer creation control 20u. When the anchor 20c clicks the question and answer creation control 20u, the social terminal 20a may display a question and answer addition page 20x which is an anchor custom question and answer page, i.e., the anchor 20c may edit the questions of interest to itself in the question and answer addition page 20x, as shown in fig. 2, the question item "who is you favorite female star? The anchor 20c may also edit its own interested options in the question and answer add page 20 x. It can be appreciated that the self-edited question-answer item of the anchor 20c needs to be checked by the live application background, and in the case that the check passes, the anchor 20c can issue the self-edited question-answer item.

If the anchor 20c clicks the question and answer selection control 20v in the question and answer adding page 20j, the social terminal 20a may display candidate question and answer items related to the food, and similarly, may provide the anchor 20c with candidate question and answer items related to the game.

Referring back to fig. 3, when the anchor clicks on item 20k, i.e., item "do empty stomach can drink milk? "upon" the social terminal 20a responds to the triggering operation for the item, a question and answer adding page 30g may be displayed, wherein the question and answer adding page 30g mainly displays a question and answer item detail area for the item 20k and a question and answer editing control. As shown in question and answer add page 30g, the question and answer item detail area may include a question item to be issued (i.e., item "do empty may drink milk. Wherein, the option items to be published may include the option item 30h (i.e., item "ok"), the option item 30i (i.e., item "not ok"), and the option item 30j (i.e., item "view") shown in fig. 3; the answer paraphrasing item to be issued is as paraphrasing item 30k illustrated in fig. 3, i.e., item "for lactose intolerant people, empty stomach does not suggest milk consumption, as it may cause discomfort to the abdomen. The question and answer editing controls described above may include the editing control 30m and the adding control 30n in fig. 3.

From the added control 30a in fig. 3, it can be known that the item "do empty can drink milk? "the item status is the item status to be issued, so fig. 3 is a description taking as an example the display of the question and answer item detail area and the question and answer editing control for the item to be issued in response to the trigger operation for the item to be issued. It can be understood that, the specific process of displaying the question item detail area and the question edit control for the candidate hot question item in response to the triggering operation for the candidate hot question item (the item state is the candidate item state herein) is implemented, and the specific process of displaying the question item detail area and the question edit control for the candidate hot question item in response to the triggering operation for the hot question item to be issued is consistent with the specific process of displaying the question item detail area and the question edit control for the hot question item to be issued, so that the specific process of displaying the question item detail area and the question edit control for the candidate hot question item in response to the triggering operation for the candidate hot question item is not repeated herein.

Referring back to fig. 3, when the anchor clicks the edit control 30m, the social terminal 20a activates the custom function of the question and answer item detail area in response to the trigger operation for the edit control 30m, at this time, the anchor may edit the question item, the option item, or the answer paraphrase item according to its own needs. As shown in question and answer add page 30p, the anchor edits the question and answer item to be issued, and as the anchor considers that milk cannot be drunk on an empty stomach for lactose intolerant people, the anchor will be uncomfortable for the bellies, so the item "can milk be drunk on an empty stomach? The answer to "is adjusted from the third option item set by the system, i.e., item" watch "to the second option item, i.e., item" not possible ".

In summary, the social terminal 20a may determine the updated to-be-issued heat question item, the updated to-be-issued heat option item, and the updated to-be-issued answer paraphrase item as to-be-issued heat question and answer items, and display the to-be-issued heat question and answer items on the screen.

It can be understood that the updated question-answer item needs to be checked by the live application background, and after the check is passed, the anchor can issue the updated question-answer item.

From the added control 30a in fig. 3, it can be known that the item "do empty can drink milk? "the item status is the item status to be issued," fig. 3 is a description taking as an example the activation of the custom function of the detail area of the question-answer item in response to the triggering operation of the edit control 30m for the item to be issued, the update of the item to be issued, the item to be issued with the answer definition based on the custom function. It can be understood that, the specific process of activating the customization function of the question and answer item detail area in response to the triggering operation for the question and answer editing control and updating the candidate hot item (here, the candidate hot item whose item state is the candidate item state), the candidate hot item and the candidate answer paraphrasing item based on the customization function is consistent with the specific process described above, so that the specific process of activating the customization function of the question and answer item detail area in response to the triggering operation for the question and answer editing control and updating the candidate hot item, the candidate hot item and the candidate answer paraphrasing item based on the customization function will not be repeated herein.

Referring back to fig. 3, when the anchor clicks the add control 30n, the social terminal 20a activates the custom function of the question and answer item detail area in response to the trigger operation for the add control 30n, and adds a new option item box for the anchor, lets the anchor edit a new option item based on the new option item box, and supports the anchor to set the new option item to be "empty can drink milk" for the item? "answer item.

Optionally, responding to a triggering operation aiming at the question and answer adding control to generate a question and answer acquisition request; the at least one candidate question item includes a first hot question item and a second hot question item; based on the question and answer acquisition request, displaying a first hot question and answer item at a first moment; determining a second moment according to the heartbeat trigger period and the first moment; the time interval between the first time and the second time is equal to the heartbeat trigger period, and the second time is later than the first time; if the second hot question and answer item acquired at the second moment is the same as the first hot question and answer item, continuing to display the first hot question and answer item; and if the second hot question and answer item is different from the first hot question and answer item, switching the first hot question and answer item to the second hot question and answer item, and displaying the second hot question and answer item.

Referring to fig. 5, fig. 5 is a schematic view of a scenario of data processing based on a social application according to an embodiment of the present application. As shown in fig. 5, at 9 points 11 minutes and 26 seconds, the anchor clicks the question and answer adding control 50c in the question and answer adding page 50b, and the social terminal 50a may generate a question and answer acquisition request in response to a trigger operation for the question and answer adding control 50 c. The social terminal 50a acquires the first hot question item and the first question item according to the question acquisition request, and displays the first hot question item 50f and the first question item 50g at a first time 50d (9 th 11 minutes 27 seconds as shown in fig. 5). As shown in the question and answer add page 50e, the first hot question and answer item 50f may include two candidate hot question and answer items, respectively, item "do it empty to drink milk? Is the "item" Piao "worth the bean cotyledon 9 minutes? "; the first question-answering item 50g may include two candidate question-answering items, respectively, item "is the young person stressed by this generation big? What is you expecting in the national celebration grade institute line, "and project? ".

In the embodiment of the present application, after the anchor clicks the question and answer adding control 50c, the social terminal 50a continuously obtains the network media data based on the heartbeat trigger period to generate a new candidate hot question and answer item, so the social terminal 50a can continuously refresh the question and answer adding page based on the heartbeat trigger period and display the latest candidate hot question and answer item, which is specifically described below.

Referring to fig. 5 again, in the embodiment of the present application, assuming that the heartbeat trigger period is equal to 2 seconds, according to the first time 50d and the heartbeat trigger period, it is known that the second time is 9 points 11 minutes 29 seconds. Acquiring a second hot question and answer item 50h at a second moment, comparing the first hot question and answer item 50f with the second hot question and answer item 50h, and if the second hot question and answer item 50h is the same as the first hot question and answer item 50f, continuing to display the first hot question and answer item 50f at the second moment by the social terminal 50 a; if the second hot question item 50h is different from the first hot question item 50f, the first hot question item 50f is switched to the second hot question item 50h, and as shown in a question adding page 50i in fig. 5, the social terminal 50a displays the second hot question item 50h, where the second hot question item 50h includes two candidate hot question items, namely, the item "empty can drink milk? Is the "and project" college entrance examination regime should be changed? ".

Step S102, a release control is displayed in response to an adding operation for at least one candidate question and answer item.

Specifically, the at least one candidate question item includes a target candidate question item; responding to the triggering operation of the candidate adding control aiming at the target candidate question-answering item, and switching the candidate adding control into an added control; the added control is used for representing that the target candidate question-answer item is the added candidate question-answer item; responding to the triggering operation aiming at the question and answer return control, and displaying a list of to-be-issued question and answer items; the to-be-issued question and answer item list comprises the added candidate question and answer items and the issued controls corresponding to the added candidate question and answer items.

The specific process of switching the candidate adding control to the added control in response to the triggering operation of the candidate adding control for the target candidate question-answering item is implemented, please refer to the description in the embodiment corresponding to fig. 3, and the details are not repeated here.

In response to a triggering operation for the question and answer return control, a list of question and answer items to be issued is displayed, please refer to fig. 6, and fig. 6 is a schematic view of a social application-based data processing scenario provided in an embodiment of the present application. As shown in fig. 6, the social terminal 60a displays a to-be-issued question-and-answer list 60b, and the to-be-issued question-and-answer list 60b includes two to-be-issued heat question-and-answer items, which are respectively the items "do empty can drink milk? Is the "item" Piao "worth the bean cotyledon 9 minutes? The to-be-issued question and answer list 60b also includes an issue control 60d and an issue control 60e. It will be appreciated that the anchor may add multiple target candidate question-answer items at a time, or may add one target candidate question-answer item at a time.

It will be appreciated that in the embodiment of the present application, describing the example in which the to-be-issued question-and-answer list 60b includes to-be-issued question-and-answer items (i.e., target candidate question-and-answer items, which may also be understood as candidate question-and-answer items, because they have not been issued at this time), in a practical application scenario, the anchor may choose to add candidate question-and-answer items, such as the item "is the young person of the generation great? ", the to-be-issued question list 60b may include candidate question items, i.e., the target candidate question item may also be a candidate question item.

Referring back to fig. 6, if the anchor clicks on the item "do empty can drink milk? The social terminal 60a may display the question and answer add page 60f, i.e., the social terminal 60a expands to display "do it empty to drink milk" for the item? "is the question and answer item detail area, in this page, the anchor can clearly know the item" is milk can empty? "candidate warmth option item, which may be the item" ok "shown in fig. 6, the item" not ok "and the item" see the situation ", and candidate answer paraphrase item, which is the item" for lactose intolerant person, as shown in fig. 6, the empty stomach does not suggest milk drinking because of causing discomfort to the abdomen ".

Step S103, responding to the triggering operation for the release control, and releasing the interactive service for the added candidate question-answer items.

Specifically, the added candidate question-answer items include candidate hotness question items and candidate hotness option items; responding to triggering operation aiming at the release control, displaying a hot question-answering sub-page, and displaying the interactive service of the added candidate question-answering items through the hot question-answering sub-page; if the triggering operation aiming at the hotness question sub-page is responded, displaying a hotness interaction detail sub-page; the hot interaction detail subpage comprises candidate hot title items, candidate hot option items and interaction data corresponding to the candidate hot option items.

Referring back to fig. 6, when the host clicks on the posting control 60d, the social terminal 60a will respond to the triggering operation for the posting control 60d, and then a hot question sub-page 60h may be displayed in the live page 60g, and the hot question sub-page 60h may display the posted hot question item, i.e., item "do empty can drink milk? "is the anchor available to the audience for the project" empty can drink milk "through the hot quiz sub-page 60 h? "interaction, including a question with the viewer of whether the empty stomach can drink milk, etc. It will be appreciated that when the anchor clicks the distribution control 60d in the question and answer adding page 60f, the user terminal corresponding to the viewing user in the live room may also display the hot question and answer page 60h.

When the anchor clicks on the hot question-answer page 60h, the social terminal 60a may expand and display content included in the hot question-answer page 60h, such as a hot interaction details sub-page 60i; as shown in fig. 6, the hotness interaction details sub-page 60i may display candidate hotness topic items (i.e., item "do empty can drink milk. The interactive data may include the number of audience members participating in the voting, such as 2 thousands of people shown in the hotness interactive details sub-page 60i, and may further include the number of votes of each option item, according to the hotness interactive details sub-page 60i, the host may know that about 1000 people select the first option item, i.e., item "ok", about 7000 people select the second option item, i.e., item "not ok", about 1 ten thousand 2 thousands of people select the third option item, i.e., item "watch", and the answer set by the host is watch.

In this case, the candidate popularity item may be regarded as a distribution popularity item, and the candidate popularity item may be regarded as a distribution popularity item.

In the embodiment of the application, at least one candidate question-answer item can be displayed by responding to the triggering operation of the question-answer adding control, wherein the at least one candidate question-answer item is a question-answer item generated based on network media data; further, in response to an add operation for at least one candidate question-answer item, a release control may be displayed; further, in response to a triggering operation for the publishing control, an interactive service for the added candidate question and answer item can be published in the social application. According to the method and the device, the display effect of the social application can be enriched, the data interaction efficiency can be improved, and the accuracy of the displayed question-answer interaction data can be guaranteed because candidate question-answer items can be automatically generated. Furthermore, the method and the device can enrich the live broadcast answer question library according to the network media data, help the host to drive audience of the live broadcast room to discuss topics through questions and answers of hot topics, improve interaction effect of the live broadcast room, and further promote active retention of the live broadcast room.

Further, referring to fig. 7, fig. 7 is a flow chart of a data processing method based on social applications according to an embodiment of the present application. As shown in fig. 7, the data processing process may include the following steps S1011-S1014, and steps S1011-S1014 are one embodiment of step S101 in the embodiment corresponding to fig. 4.

Step S1011, in response to a trigger operation for the question and answer addition control, a question and answer acquisition request is generated.

The social terminal may generate a question-answer obtaining request in response to the triggering operation of the question-answer adding control, and the specific implementation process is described in the embodiment corresponding to fig. 5, which is not described herein.

Step S1012, acquiring the network media data based on the question-answer acquisition request, and inputting the network media data into the popularity recommendation model.

Specifically, referring to fig. 8, fig. 8 is a schematic view of a scenario of data processing based on social applications according to an embodiment of the present application. As shown in fig. 8, the social terminal 80a generates a question-answer acquisition request in response to the triggering operation for the question-answer adding control 80b, and acquires the network media data 80c in the internet based on the question-answer acquisition request, it may be understood that the network media data 80c may be news, may be a novel or a television show, and the specific data type of the network media data 80c is not limited in the embodiment of the present application. Meanwhile, the present application does not limit the number of network media data 80c, and may be one or more network media data.

The social terminal 80a inputs the acquired network media data 80c into a popularity recommendation model 80d, as shown in fig. 8, the popularity recommendation model 80d may include a base layer, a recommendation layer, and a ranking layer.

In step S1013, real-time popularity media data is determined from the network media data based on the popularity recommendation model.

Specifically, the heat recommendation model comprises a base layer, a recommendation layer and a sequencing layer; extracting real-time characteristics of network media data and user characteristics of user historical behavior data based on a base layer, and inputting the real-time characteristics and the user characteristics into a recommendation layer; the user historical behavior data refers to user behavior data which has performed interactive operation on the network media data; identifying real-time features and user features based on the recommendation layer, determining candidate hot media data from the network media data according to the identification result, and inputting the candidate hot media data into the sequencing layer; and sorting the candidate hot media data based on the sorting layer, and determining the real-time hot media data from the candidate hot media data according to the sorting result.

The specific implementation method for extracting the real-time characteristics of the network media data and the user characteristics of the user history behavior data based on the base layer can comprise the following steps: performing semantic recognition processing on the network media data based on a semantic recognition component in the base layer to obtain real-time characteristics of the network media data; and performing behavior recognition processing on the user historical behavior data based on a user recognition component in the base layer to obtain user characteristics of the user historical behavior data.

The recommendation layer comprises a user recommendation layer, an article recommendation layer and a content recommendation layer; the candidate popularity media data comprises user popularity media data, article popularity media data and content popularity media data; based on the recommendation layer, identifying real-time characteristics and user characteristics, and determining candidate hot media data from the network media data according to the identification result, the specific implementation method can comprise the following steps: identifying user characteristics based on a user recommendation layer, scoring the network media data according to a user identification result, and determining the user popularity media data from the network media data according to the user popularity score; identifying user characteristics based on the item recommendation layer, carrying out item popularity scoring on the network media data according to an item identification result, and determining item popularity media data from the network media data according to the item popularity scoring; identifying real-time characteristics based on the content recommendation layer, and performing content labeling processing on the network media data according to the content identification result to obtain a content labeling result; and determining content hot media data from the network media data according to the user characteristics and the content labeling result.

The ordering layer comprises a data de-duplication layer and a data ordering layer; the specific implementation method for sorting the candidate heat media data based on the sorting layer can comprise the following steps: performing de-duplication processing on the user heat media data, the article heat media data and the content heat media data based on the data de-duplication layer to obtain target candidate heat media data; and inputting the target candidate hot media data into a data sorting layer, and sorting the target candidate hot media data based on the data sorting layer.

Optionally, the data sorting layer includes a forwarding data sorting layer, a comment data sorting layer, and a forward data sorting layer; the specific implementation method for inputting the target candidate hotness media data into the data sorting layer and sorting the target candidate hotness media data based on the data sorting layer can comprise the following steps: respectively inputting the target candidate hot media data into a forwarding data sorting layer, a comment data sorting layer and a forward data sorting layer; determining forwarding probability of the target candidate hot media data based on the forwarding data ordering layer, determining comment probability of the target candidate hot media data based on the comment data ordering layer, and determining forward probability of the target candidate hot media data based on the forward data ordering layer; weighting and fusing the forwarding probability, the comment probability and the forward probability to obtain the fusion probability of the target candidate heat media data; and sorting the target candidate hot media data based on the fusion probability.

Referring to fig. 8 again, the social terminal 80a inputs the network media data 80c into a base layer, where the base layer may include a semantic recognition component and a user recognition component, and performs semantic recognition processing on the network media data 80c based on the semantic recognition component in the base layer to obtain real-time features 80f of the network media data; and performing behavior recognition processing on the user historical behavior data based on the user recognition component in the base layer to obtain user characteristics 80e of the user historical behavior data. The above-described user history behavior data may include a user's history browsing amount (click amount, forward amount, comment amount, praise, etc.), user's influence, user type, and the like. It is to be appreciated that the semantic recognition component as well as the user recognition component can be a pre-trained system.

The social terminal 80a inputs the real-time feature 80f and the user feature 80e into a recommendation layer, identifies the real-time feature 80f and the user feature 80e based on the recommendation layer, performs a popularity score on the network media data 80c according to the identification result, and finally determines candidate popularity media data 80g from the network media data 80c according to the popularity score. The candidate popularity media data 80g may include, among other things, user popularity media data determined based on the user recommendation layer, item popularity media data determined based on the item recommendation layer, and content popularity media data determined based on the content recommendation layer. It can be understood that the recommendation algorithm adopted in the application is a recommendation algorithm based on a user (i.e. a user recommendation layer), a recommendation algorithm based on an article (i.e. an article recommendation layer) and a recommendation algorithm based on a content (i.e. a content recommendation layer), but in an actual application scenario, the recommendation algorithm can be adopted according to actual situations, and the recommendation algorithm in the recommendation layer is not limited in the application. It is understood that the user recommendation layer, the item recommendation layer, and the content recommendation layer may be pre-trained systems.

It is to be understood that the candidate heat media data 80g may be one or more candidate heat media data, and the number of the candidate heat media data 80g is not limited in the present application, and may be set according to a scene in actual application.

Referring to fig. 8 again, the social terminal 80a inputs the candidate popularity media data 80g into a ranking layer, which includes a data deduplication layer and a data ranking layer; the candidate hot media data 80g (including user hot media data, item hot media data, and content hot media data) is de-duplicated based on the data de-duplication layer, i.e., one and the same candidate hot media data is retained. If the candidate hot media data 80g is a candidate hot media data, the deduplication step is skipped.

The social terminal 80a may obtain target candidate popularity media data, which may be one or more target candidate popularity media data, which should be different when there are multiple target candidate popularity media data. And inputting the target candidate hot media data into a data sorting layer, and sorting the target candidate hot media data based on the data sorting layer. The data sorting layer comprises a forwarding data sorting layer, a comment data sorting layer and a forward data sorting layer, wherein the forwarding data sorting layer mainly determines the later forwarding condition, namely the possible forwarding probability, of the target candidate hot media data according to the existing forwarding data of the target candidate hot media data; the comment data ordering layer mainly determines the later comment condition, namely the possible comment probability, of the target candidate hot media data according to the existing comment data of the target candidate hot media data; the forward data sorting layer is set for the praise data of the target candidate hot media data, and may also be referred to as a praise data sorting layer, and the praise data sorting layer determines the later praise condition of the target candidate hot media data, i.e. the possible praise probability (i.e. forward probability) mainly according to the existing praise data condition of the target candidate hot media data. The social terminal 80a performs weighted fusion on the forwarding probability, the comment probability and the forward probability to obtain fusion probability of the target candidate heat media data; and finally, sorting the target candidate hot media data based on the fusion probability. According to the ranking result, the social terminal 80a determines real-time popularity media data 80h from among the candidate popularity media data 80 g.

It is understood that the forwarding data ordering layer, the comment data ordering layer, and the forward data ordering layer may be pre-trained systems or network models.

It is to be understood that the real-time heat media data 80h may be one or more real-time heat media data, and the present application does not limit the number of the real-time heat media data 80h, and may be set according to a scene in actual application.

Step S1014, outputting candidate hot question-answering items according to the real-time hot media data.

Specifically, acquiring heat comment data of real-time heat media data, and acquiring heat keywords of the heat comment data; and outputting candidate hot question-answering items according to the hot keywords.

Referring to fig. 8 again, after determining the real-time popularity media data 80h, the social terminal 80a further obtains popularity comment data (i.e. the hot topics shown in fig. 8) of the real-time popularity media data 80h, which is not limited in the manner of obtaining the hot topics, and may be set according to the application scenario in practical application through a web crawler manner or a code manner. It is understood that the hot comment data described above may include hot topics that are most commented and discussed in the real-time hot media data 80 h.

The social terminal 80a obtains the hotness keywords of the hotness comment data (i.e., the hot keywords in fig. 8), which are not limited in the manner of obtaining the hotness keywords, and may be obtained through a supervised keyword extraction algorithm, such as a neural network model; the hotness keywords, such as a semi-supervised neural network model, can be obtained through a semi-supervised keyword extraction algorithm; hotness keywords can be obtained through an unsupervised keyword extraction algorithm, such as a keyword extraction mode based on statistical features, a keyword extraction algorithm based on a word graph model and a keyword extraction algorithm based on a topic model. In actual application, the setting of the hotness keyword acquisition algorithm can be performed according to the application scene.

Social terminal 80a generates candidate hotness question and answer items 80i based on hotness keywords, e.g., keywords of "empty stomach" and "milk", then the item "empty stomach can drink milk? The method for specifically generating the item is not limited, the process from the hotness key word to at least one candidate question-answering item can be realized based on the deep neural network, the process from the hotness key word to at least one candidate question-answering item can be realized based on a semi-supervision mode, and the method can be set according to application scenes in actual application.

The social terminal 80a displays the candidate hot topic item 80j among the candidate hot question items 80i in the question and answer addition page 80k as shown in fig. 8.

It will be appreciated that candidate question-answer items, such as the item "is the young age of the generation stressed by the person? The generation process of the candidate question-answer item 80i may be identical to the generation process of the candidate question-answer item, so that the generation process of the candidate question-answer item is not repeated. The difference between the candidate question and answer items 80i is mainly that they are generated based on real-time hot media data, have real-time performance and hot spot performance, and the candidate question and answer items may be generated based on real-time media data or common media data, and the generation source data of the candidate question and answer items is not limited here.

In the embodiment of the application, at least one candidate question-answer item can be displayed by responding to the triggering operation of the question-answer adding control, wherein the at least one candidate question-answer item is a question-answer item generated based on network media data; further, in response to an add operation for at least one candidate question-answer item, a release control may be displayed; further, in response to a triggering operation for the publishing control, an interactive service for the added candidate question and answer item can be published in the social application. According to the method and the device, the display effect of the social application can be enriched, the data interaction efficiency can be improved, and the accuracy of the displayed question-answer interaction data can be guaranteed because candidate question-answer items can be automatically generated. Furthermore, the live broadcast answering library can be enriched according to the hot real-time news (namely the real-time hot media data), and the live broadcast answering library can help a host to drive audience of a live broadcast room to discuss topics through question answering of hot topics, so that interaction effect of the live broadcast room is improved, and active retention of the live broadcast room is further improved.

Further, referring to fig. 9 in combination with the embodiments corresponding to fig. 4 and fig. 7, fig. 9 is a flow chart of a data processing method according to an embodiment of the present application. The data processing method may be executed by a computer device, which may be the social terminal or the application server 10a shown in fig. 1, so that the data processing method may also be executed by the social terminal and the application server 10a together, and in this embodiment of the present application, the method is described as an example of being executed by the social terminal and the application server 10a together. As shown in fig. 9, the data processing procedure includes the steps of:

step S201-step S203, the anchor starts live broadcast, clicks a question and answer button to enter a question and answer adding page, and the client requests a hot question and answer library from the server in real time.

Specifically, the question and answer button described above is equivalent to the question and answer adding control 20i in fig. 2, is also equivalent to the question and answer adding control 50c in fig. 5, and is also equivalent to the question and answer adding control 80b in fig. 8. The anchor clicks the question and answer button and the client displays a question and answer add page equivalent to the question and answer add page 20j in fig. 2, equivalent to the question and answer add page 50e in fig. 5, and equivalent to the question and answer add page 80k in fig. 8.

The clients described above are equivalent to the social terminal 20a in fig. 2, also equivalent to the social terminal 50a in fig. 5, and also equivalent to the social terminal 80a in fig. 8. The above-described server is equivalent to the application server 10a in fig. 1. The hot question library may be understood as the at least one candidate question item, and may include candidate hot question items and candidate question items.

The specific implementation process of step S201 to step S203 may refer to the description of step S101 in the embodiment corresponding to fig. 2, or may refer to the description of step S1011 in the embodiment corresponding to fig. 7, which is not repeated here.

Step S204-step S206, the server captures the whole-network hot news in real time, and based on the hottest topics discussed by the users in the hot news, the live question and answer questions are intelligently generated.

In particular, the whole-network news can be understood as the above network media data, and in practical application, the network media data includes, but is not limited to, whole-network hot news. The server determines hot news (which may be understood as the real-time hot media data described above) according to a recommender system algorithm (i.e., the hot recommendation model described above), and generates a live question-answer item (which may be understood as the at least one candidate question-answer item described above).

Briefly described below, the principles of determining hot topics for hot news, a recommender algorithm may include a machine learning based multi-way recall and sort strategy, and a recommendation engine from offline computing of massive large data to high concurrency online services. Therefore, the heat recommendation model is a trained model.

The popularity recommendation model is mainly divided into three parts, namely a base layer, a recommendation (recall) layer and a sequencing layer, wherein the recommendation layer is mainly responsible for generating candidate popularity media data, and the sequencing layer is responsible for personalized sequencing of results of a plurality of recommendation algorithms (namely candidate popularity media data). Referring to fig. 10, fig. 10 is a schematic structural diagram of a heat recommendation model according to an embodiment of the present application. As shown in fig. 10, the base layer provides real-time features and user features for the recommendation layer, the recommendation layer provides candidate sets (i.e., candidate hotness media data) for the ranking layer, and the ranking layer outputs the ranked recommendation results (i.e., real-time hotness media data). The base layer, recommendation layer, and ranking layer are briefly described below.

Base layer: the method is divided into two parts of content modeling and user modeling. Content modeling may be equivalent to the semantic recognition component described above, mainly semantic recognition of content (i.e., whole-web news), including topic model, entity word recognition, text classification, and picture classification, resulting in real-time features. User modeling is equivalent to the user identification component described above, and a complete portrait is created for the user, including user natural attributes (gender/age), user interests, user clusters, and relationships between users (affinity, etc.), resulting in user features.

Recommendation layer: different candidate sets are obtained through real-time features and user features, and multiple recall algorithms. The specific recall algorithm is as follows:

user-based collaborative recommendation: n users most similar to the current User X are found, and the scoring of an Item (an Item, which can be referred to as a real-time news) by the User X is estimated according to the scoring of the Item by the N users.

Item-based (Item-based) collaborative recommendation: and calculating the co-occurrence probability of different m article identifications, and taking out the article identifications which meet a certain threshold and are arranged in front as candidates of the cooperative article identifications.

Content-based collaborative recommendation: labeling the text, picture, video and other contents in the network media data through algorithms such as natural language processing, image recognition and the like; and mining interest tags of the users through the user characteristics and the content tags. Based on the matching of the content tags and the interest tags, content-based recommendation candidates are provided.

Sequencing layer: each type of recall strategy recalls candidate hot news, and the candidate hot news needs to be uniformly ordered after duplicate removal. The model used for sequencing includes logistic regression, gradient descent tree/factorizer, deep neural network, etc. The process of training the ordering framework can be roughly divided into three parts:

Feature engineering: feature preprocessing, discretization, normalization, feature combination and the like, and sample data required by a training model is generated.

Model tool: based on the sample data, different models are used for training and evaluation, and model training results are generated.

Ordering engine: and (3) importing an online model, extracting features corresponding to training results, performing feature mapping, and fusing and scoring rearrangement on the recommended candidates for multi-strategy recall by using a machine learning ordering algorithm.

Referring to fig. 11 together with the trained ranking framework, fig. 11 is a schematic diagram of a model structure of the ranking framework according to the embodiment of the present application. As shown in fig. 11, in the embodiment of the present application, 3 targets, that is, forwarding, commenting, and praise, are included, and each target uses a model, such as the estimated forwarding model, the estimated commenting model, and the estimated praise model shown in fig. 11. The training and sorting framework is to train the estimated forwarding model, the estimated comment model and the estimated praise model respectively, namely train the forwarding probability P (y) ₁ |x _u ，x _i ，x _c ) Comment probability P (y ₂ |x _u ，x _i ，x _c ) Praise probability P (y ₃ |x _u ，x _i ，x _c ) And then weighting the prediction results of the three models, and taking the weighted fusion model as a sequencing layer. Wherein x is _u Representing the user x _i Representing an article, x _c And the representative content, u, i and c are positive integers.

Wherein the weighting mode is not limited, in the embodiment of the application, linear weighting is adopted Exponentially weighted ++>For example, in actual application, setting may be performed according to an application scenario.

After the models are fused, the probability of promoting all forward behaviors is taken as a total target, and different weights are distributed to the models. The method has the advantages that different targets are respectively modeled, and the weights are adjusted by a plurality of groups of fast experiments so as to find a better solution of the weight parameters. All targets use one model, and multiple targets are considered when labeling positive samples. For example, for forwarding and praise, when labeling positive samples, different weights are given so that they are comprehensively embodied in the model target. The weights of forwarding, commenting and praying can be respectively set to 1, 2 and 5 according to the service requirement.

The method converts the multi-objective problem into a single-objective problem by giving different weights to different forward behaviors. The method has the advantages that the model is fused, a plurality of targets are considered, weights of different targets are reflected in the loss function, the model is participated in optimization solution, and the effects of the targets are balanced conveniently.

Step S207-step S208, generating answers and options of corresponding questions by linking the big data, generating a live broadcast hot spot question bank and transmitting the live broadcast hot spot question bank back to the client.

Specifically, after generating the live question and answer item (i.e., the candidate hotness item), the server links the big data to generate an answer and options of the corresponding item, the options (i.e., the candidate hotness item) are generated based on interactive comments of the user in the real-time hotness media data, and the answer is generated from the whole-network search of the big data based on the candidate hotness item. Typically, each candidate hotness item corresponds to 3-4 candidate hotness items, 1 correct answer.

Step S209-step S212, the client displays a real-time hot spot question bank and marks flames; adding hot spot question and answer questions by the anchor and publishing the questions; the client displays a related problem widget in a hosting room; the background pushes questions and answers to the user side to start interaction.

Specifically, the flame is equivalent to the heat mark 20n and the heat mark 20p in fig. 2; the related questions portlet described above is equivalent to the hot question answering page 60h in FIG. 6; the above-mentioned user terminal is equivalent to the user terminal (e.g., the user terminal 101 b) in fig. 1.

The server transmits the hot spot question bank to the client, and the client displays the real-time hot spot question bank and presents a flame icon in front of the hot spot question bank. After the host adds the hot spot question and answer questions and clicks the release button, the server pushes a popup window for the question and answer questions to the user side of the room where the host id is located, and the user can participate in question and answer interaction by clicking the popup window. At this time, the room at the anchor end will also show a small window for the relevant questions.

Further, referring to fig. 12, fig. 12 is a schematic structural diagram of a social application-based data processing apparatus according to an embodiment of the present application. The data processing means may be a computer program (comprising program code) running in a computer device, for example the data processing means is an application software; the device can be used for executing corresponding steps in the method provided by the embodiment of the application. As shown in fig. 12, the data processing apparatus 1 may include: a first response module 11, a second response module 12 and a third response module 13.

A first response module 11, configured to display at least one candidate question-answer item in response to a trigger operation for a question-answer adding control, where the at least one candidate question-answer item is generated based on network media data;

a second response module 12, configured to display a release control in response to an add operation for at least one candidate question-answer item;

and the third response module 13 is used for responding to the triggering operation for the release control and releasing the interactive service for the added candidate question-answer items.

The specific functional implementation manner of the first response module 11, the second response module 12, and the third response module 13 may be referred to step S101-step S103 in the corresponding embodiment of fig. 4, which is not described herein.

Referring again to fig. 12, the at least one candidate question-answer item includes a candidate hot question-answer item;

the first response module 11 may include: a first display unit 111, a second display unit 112, an activation function unit 113, and a first determination unit 114.

A first display unit 111 for displaying a popularity identifier for characterizing that a question-answer type is a popularity question-answer type, and a candidate popularity topic item in response to a trigger operation for a question-answer addition control; the question-answer type of the candidate hot question item is a hot question-answer type;

a second display unit 112 for displaying a question-answer item detail area and a question-answer editing control for the candidate hot item in response to a trigger operation for the candidate hot item; the question and answer item detail area comprises a candidate heat question item, a candidate heat option item and a candidate answer paraphrasing item;

an activation function unit 113, configured to activate a custom function of a question-answer item detail area in response to a trigger operation for the question-answer editing control, and update a candidate hotness item, and a candidate answer paraphrasing item based on the custom function;

the first determining unit 114 is configured to determine the updated candidate hot question item, the updated candidate hot option item, and the updated candidate answer paraphrase item as candidate hot question items, and display the candidate hot question items.

The specific functional implementation manners of the first display unit 111, the second display unit 112, the activation function unit 113, and the first determining unit 114 may be referred to the step S101 in the corresponding embodiment of fig. 4, and will not be described herein.

Referring again to fig. 12, the at least one candidate question item includes a target candidate question item;

the second response module 12 may include: the first response unit 121 and the second response unit 122.

A first response unit 121, configured to switch a candidate adding control to an added control in response to a triggering operation of the candidate adding control for a target candidate question-answer item; the added control is used for representing that the target candidate question-answer item is the added candidate question-answer item;

a second response unit 122, configured to respond to a triggering operation for the question-answer return control, and display a list of to-be-issued question-answer items; the to-be-issued question and answer item list comprises the object candidate question and answer items and the issuing controls corresponding to the object candidate question and answer items.

The specific functional implementation manner of the first response unit 121 and the second response unit 122 may refer to step S102 in the corresponding embodiment of fig. 4, which is not described herein.

Referring to fig. 12 again, the added candidate question-answer items include a candidate popularity item and a candidate popularity item;

The third response module 13 may include: the third display unit 131 and the fourth display unit 132.

A third display unit 131, configured to display a hot question-answer page in response to a trigger operation for the release control, and display, through the hot question-answer page, an interactive service of the added candidate question-answer item;

a fourth display unit 132, configured to display a hotness interaction details sub-page if a trigger operation for the hotness question sub-page is responded; the hot interaction detail subpage comprises candidate hot title items, candidate hot option items and interaction data corresponding to the candidate hot option items.

The specific functional implementation manner of the third display unit 131 and the fourth display unit 132 may refer to step S103 in the corresponding embodiment of fig. 4, which is not described herein.

Referring again to fig. 12, the first response module 11 may include: the first generation unit 115 and the fifth display unit 116.

A first generating unit 115 configured to generate a question-answer acquisition request in response to a trigger operation for a question-answer addition control;

and a fifth display unit 116 for displaying at least one candidate question-answer item based on the heartbeat trigger period and the question-answer acquisition request.

The specific functional implementation manner of the first generating unit 115 and the fifth displaying unit 116 may refer to step S101 in the corresponding embodiment of fig. 4, and will not be described herein.

Referring again to fig. 12, the at least one candidate question item includes a first hot question item and a second hot question item;

the fifth display unit 116 may include: a first display subunit 1161, a first determination subunit 1162, a second display subunit 1163, and a third display subunit 1164.

A first display subunit 1161, configured to display a first heat question and answer item at a first time based on the question and answer acquisition request;

a first determining subunit 1162, configured to determine a second time according to the heartbeat trigger period and the first time; the time interval between the first time and the second time is equal to the heartbeat trigger period, and the second time is later than the first time;

the second display subunit 1163 is configured to continuously display the first hot question and answer item if the second hot question and answer item acquired at the second moment is the same as the first hot question and answer item;

and a third display subunit 1164, configured to switch the first hot question and answer item to the second hot question and answer item if the second hot question and answer item is different from the first hot question and answer item, and display the second hot question and answer item.

The specific functional implementation manner of the first display subunit 1161, the first determining subunit 1162, the second display subunit 1163, and the third display subunit 1164 may refer to step S101 in the corresponding embodiment of fig. 4, and will not be described herein.

the first response module 11 may include: a second generation unit 117, a second determination unit 118, and an output item unit 119.

A second generating unit 117 for generating a question-answer acquisition request in response to a trigger operation for a question-answer addition control;

the second generating unit 117 is further configured to acquire network media data based on the question-answer acquisition request, and input the network media data into the popularity recommendation model;

a second determining unit 118, configured to determine real-time popularity media data from the network media data based on the popularity recommendation model;

an output item unit 119 for outputting candidate popularity question-answer items according to the real-time popularity media data.

The specific functional implementation manners of the second generating unit 117, the second determining unit 118, and the output item unit 119 may be referred to as step S1011-step S1014 in the corresponding embodiment of fig. 7, which are not described herein.

Referring to fig. 12, the popularity recommendation model includes a base layer, a recommendation layer, and a ranking layer;

the second determining unit 118 may include: the extraction feature subunit 1181, the second determination subunit 1182, and the third determination subunit 1183.

An extracted feature subunit 1181, configured to extract real-time features of the network media data and user features of the user history behavior data based on the base layer, and input the real-time features and the user features into the recommendation layer; the user historical behavior data refers to user behavior data which has performed interactive operation on the network media data;

a second determining subunit 1182, configured to identify real-time features and user features based on the recommendation layer, determine candidate popularity media data from the network media data according to the identification result, and input the candidate popularity media data into the ranking layer;

the third determining subunit 1183 is configured to perform a ranking process on the candidate popularity media data based on the ranking layer, and determine real-time popularity media data from the candidate popularity media data according to the ranking result.

The specific functional implementation manner of the extracted feature subunit 1181, the second determining subunit 1182, and the third determining subunit 1183 may refer to step S1013 in the corresponding embodiment of fig. 7, and are not described herein.

Referring again to fig. 12, the extracted features subunit 1181 may include: a first processing subunit 11811 and a second processing subunit 11812.

A first processing subunit 11811, configured to perform semantic recognition processing on the network media data based on the semantic recognition component in the base layer, so as to obtain real-time features of the network media data;

the second processing subunit 11812 is configured to perform behavior recognition processing on the user historical behavior data based on the user recognition component in the base layer, so as to obtain the user characteristics of the user historical behavior data.

The specific functional implementation manner of the first processing subunit 11811 and the second processing subunit 11812 may refer to step S1013 in the corresponding embodiment of fig. 7, and will not be described herein.

Referring to fig. 12 again, the recommendation layers include a user recommendation layer, an item recommendation layer, and a content recommendation layer; the candidate popularity media data comprises user popularity media data, article popularity media data and content popularity media data;

the second determination subunit 1182 may include: a first identification subunit 11821, a second identification subunit 11822, and a third identification subunit 11823.

A first identifying subunit 11821, configured to identify a user feature based on the user recommendation layer, perform user popularity scoring on the network media data according to the user identification result, and determine user popularity media data from the network media data according to the user popularity scoring;

A second identifying subunit 11822, configured to identify the user feature based on the item recommendation layer, score the item popularity of the network media data according to the item identification result, and determine the item popularity media data from the network media data according to the item popularity score;

the third identifying subunit 11823 is configured to identify real-time features based on the content recommendation layer, and perform content labeling processing on the network media data according to the content identification result to obtain a content labeling result;

the third identifying subunit 11823 is further configured to determine content hot media data from the network media data according to the user characteristics and the content labeling result.

The specific functional implementation manner of the first recognition subunit 11821, the second recognition subunit 11822 and the third recognition subunit 11823 may refer to step S1013 in the corresponding embodiment of fig. 7, and will not be described herein.

Referring to fig. 12 again, the ordering layer includes a data de-duplication layer and a data ordering layer;

the third determination subunit 1183 may include: third processing subunit 11831 and fourth processing subunit 11832.

The third processing subunit 11831 is configured to perform deduplication processing on the user popularity media data, the item popularity media data, and the content popularity media data based on the data deduplication layer, to obtain target candidate popularity media data;

The fourth processing subunit 11832 is configured to input the target candidate popularity media data into the data ordering layer, and perform an ordering process on the target candidate popularity media data based on the data ordering layer.

The specific functional implementation manner of the third processing subunit 11831 and the fourth processing subunit 11832 may refer to step S1013 in the corresponding embodiment of fig. 7, and will not be described herein.

Referring to fig. 12 again, the data sorting layer includes a forwarding data sorting layer, a comment data sorting layer, and a forward data sorting layer;

the fourth processing subunit 11832 is specifically configured to input the target candidate popularity media data into the forwarding data ordering layer, the comment data ordering layer, and the forward data ordering layer, respectively;

the fourth processing subunit 11832 is further specifically configured to determine, based on the forwarding data ordering layer, a forwarding probability of the target candidate hot media data, determine, based on the comment data ordering layer, a comment probability of the target candidate hot media data, and determine, based on the forward data ordering layer, a forward probability of the target candidate hot media data;

the fourth processing subunit 11832 is further specifically configured to perform weighted fusion on the forwarding probability, the comment probability, and the forward probability, so as to obtain a fusion probability of the target candidate heat media data;

The fourth processing subunit 11832 is further specifically configured to perform a ranking process on the target candidate hot media data based on the fusion probability.

The specific functional implementation of the fourth processing subunit 11832 may refer to step S1013 in the corresponding embodiment of fig. 7, which is not described herein.

Referring again to fig. 12, the output item unit 119 may include: the get heat subunit 1191 and the output item subunit 1192.

An acquiring heat subunit 1191, configured to acquire heat comment data of the real-time heat media data, and acquire a heat keyword of the heat comment data;

an output item subunit 1192 is configured to output at least one candidate question and answer item according to the hotness key.

The specific function implementation manner of the heat obtaining subunit 1191 and the output item subunit 1192 may refer to step S1014 in the corresponding embodiment of fig. 7, which is not described herein.

Further, referring to fig. 13, fig. 13 is a schematic structural diagram of a computer device according to an embodiment of the present application. As shown in fig. 13, the computer device 1000 may include: at least one processor 1001, such as a CPU, at least one network interface 1004, a user interface 1003, a memory 1005, at least one communication bus 1002. Wherein the communication bus 1002 is used to enable connected communication between these components. The user interface 1003 may include a Display (Display), a Keyboard (Keyboard), and the network interface 1004 may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface), among others. The memory 1005 may be a high-speed RAM memory or a non-volatile memory (non-volatile memory), such as at least one disk memory. The memory 1005 may also optionally be at least one storage device located remotely from the aforementioned processor 1001. As shown in fig. 13, the memory 1005, which is one type of computer storage medium, may include an operating system, a network communication module, a user interface module, and a device control application.

In the computer device 1000 shown in FIG. 13, the network interface 1004 may provide network communication functions; while user interface 1003 is primarily used as an interface for providing input to a user; and the processor 1001 may be used to invoke a device control application stored in the memory 1005 to implement:

It should be understood that the computer device 1000 described in the embodiments of the present application may perform the description of the data processing method in the embodiments corresponding to fig. 4, fig. 7, and fig. 9, and may also perform the description of the data processing apparatus 1 in the embodiments corresponding to fig. 12, which are not described herein. In addition, the description of the beneficial effects of the same method is omitted.

The embodiment of the present application further provides a computer readable storage medium, where the computer readable storage medium stores a computer program, where the computer program includes program instructions, where the program instructions, when executed by a processor, implement a social application-based data processing method provided in each step of fig. 4, fig. 7, and fig. 9, and specifically refer to an implementation manner provided in each step of fig. 4, fig. 7, and fig. 9, which is not described herein again.

The computer readable storage medium may be the data processing apparatus provided in any one of the foregoing embodiments or an internal storage unit of the computer device, for example, a hard disk or a memory of the computer device. The computer readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash card (flash card) or the like, which are provided on the computer device. Further, the computer-readable storage medium may also include both internal storage units and external storage devices of the computer device. The computer-readable storage medium is used to store the computer program and other programs and data required by the computer device. The computer-readable storage medium may also be used to temporarily store data that has been output or is to be output.

In one aspect, the present application provides a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the computer device can execute the description of the data processing method in the embodiments corresponding to fig. 4, fig. 7, and fig. 9, which are not described herein. In addition, the description of the beneficial effects of the same method is omitted.

The terms first, second and the like in the description and in the claims and drawings of the embodiments of the present application are used for distinguishing between different objects and not for describing a particular sequential order. Furthermore, the term "include" and any variations thereof is intended to cover a non-exclusive inclusion. For example, a process, method, apparatus, article, or device that comprises a list of steps or elements is not limited to the list of steps or modules but may, in the alternative, include other steps or modules not listed or inherent to such process, method, apparatus, article, or device.

Those of ordinary skill in the art will appreciate that the elements and algorithm steps described in connection with the embodiments disclosed herein may be embodied in electronic hardware, in computer software, or in a combination of the two, and that the elements and steps of the examples have been generally described in terms of function in the foregoing description to clearly illustrate the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

The methods and related devices provided in the embodiments of the present application are described with reference to the method flowcharts and/or structure diagrams provided in the embodiments of the present application, and each flowchart and/or block of the method flowcharts and/or structure diagrams may be implemented by computer program instructions, and combinations of flowcharts and/or blocks in the flowchart and/or block diagrams. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or structural diagram block or blocks. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or structures.

The foregoing disclosure is only illustrative of the preferred embodiments of the present application and is not intended to limit the scope of the claims herein, as the equivalent of the claims herein shall be construed to fall within the scope of the claims herein.

Claims

1. A method for processing data based on a social application, comprising:

responding to a triggering operation aiming at a question and answer adding control, and generating a question and answer acquisition request;

acquiring network media data based on the question-answer acquisition request, and inputting the network media data into a hotness recommendation model; the heat recommendation model comprises a base layer, a recommendation layer and a sequencing layer;

extracting real-time characteristics of the network media data and user characteristics of user history behavior data based on the base layer, and inputting the real-time characteristics and the user characteristics into the recommendation layer; the user history behavior data refers to user behavior data which has performed interactive operation on the network media data;

identifying the real-time features and the user features based on the recommendation layer, determining candidate popularity media data from the network media data according to the identification result, and inputting the candidate popularity media data into the ranking layer;

Sorting the candidate hot media data based on the sorting layer, and determining real-time hot media data from the candidate hot media data according to a sorting result;

outputting candidate hot question-answering items according to the real-time hot media data, and displaying at least one candidate question-answering item comprising the candidate hot question-answering items;

responding to the adding operation aiming at the at least one candidate question-answer item, and displaying a release control;

and responding to the triggering operation of the release control, and releasing the interactive service of the added candidate question-answer item.

2. The method of claim 1, wherein the outputting candidate hot question items from the real-time hot media data, displaying at least one candidate question item that includes the candidate hot question items, comprises:

displaying a popularity identifier for representing that the question-answer type is a popularity question-answer type and candidate popularity topic items according to the real-time popularity media data; the question and answer type of the candidate hot question item is the hot question and answer type;

responding to the triggering operation of the candidate hot question item, and displaying a question and answer item detail area and a question and answer editing control of the candidate hot question item; the question and answer item detail area comprises the candidate hotness question item, the candidate hotness option item and the candidate answer paraphrasing item;

Responding to the triggering operation aiming at the question and answer editing control, activating a self-defining function of the question and answer item detail area, and updating the candidate hot question item, the candidate hot option item and the candidate answer paraphrasing item based on the self-defining function;

and determining the updated candidate hot question item, the updated candidate hot option item and the updated candidate answer paraphrasing item as candidate hot question and answer items, and displaying at least one candidate question and answer item comprising the candidate hot question and answer items.

3. The method of claim 1, wherein the at least one candidate question-answer item comprises a target candidate question-answer item;

the response to the adding operation of the at least one candidate question-answer item displays a release control, including:

responding to the triggering operation of the candidate adding control aiming at the target candidate question-answering item, and switching the candidate adding control into an added control; the added control is used for representing the target candidate question-answer item as the added candidate question-answer item;

responding to the triggering operation aiming at the question and answer return control, and displaying a list of to-be-issued question and answer items; the to-be-published question and answer item list comprises the added candidate question and answer items and the publishing control corresponding to the added candidate question and answer items.

4. The method of claim 1, wherein the added candidate question-answer items include a candidate hotness question item and a candidate hotness option item;

the response to the triggering operation of the release control releases the interactive service for the added candidate question-answer items, and the method comprises the following steps:

responding to the triggering operation aiming at the release control, displaying a hot question-answering sub-page, and displaying the interactive service of the added candidate question-answering item through the hot question-answering sub-page;

if the triggering operation aiming at the hot question-answer sub-page is responded, displaying a hot interaction detail sub-page; the hotness interaction detail sub-page comprises the candidate hotness question item, the candidate hotness option item and interaction data corresponding to the candidate hotness option item.

5. The method of claim 1, wherein the displaying at least one candidate question-answer item that includes the candidate hot question-answer item comprises:

and displaying the at least one candidate question-answer item based on the heartbeat trigger period.

6. The method of claim 5, wherein the at least one candidate question item comprises a first hot question item and a second hot question item;

The displaying at least one candidate question-answer item including the candidate hot question-answer item based on the heartbeat trigger period includes:

displaying the first hot question and answer item at a first moment;

determining a second moment according to the heartbeat trigger period and the first moment; the time interval between the first time and the second time is equal to the heartbeat trigger period, and the second time is later than the first time;

if the second hot question and answer item acquired at the second moment is the same as the first hot question and answer item, continuing to display the first hot question and answer item;

and if the second hot question and answer item is different from the first hot question and answer item, switching the first hot question and answer item to the second hot question and answer item, and displaying the second hot question and answer item.

7. The method of claim 1, wherein the extracting real-time features of the network media data and user features of user historical behavior data based on the base layer comprises:

performing semantic recognition processing on the network media data based on a semantic recognition component in the base layer to obtain the real-time characteristics of the network media data;

And performing behavior recognition processing on the user historical behavior data based on a user recognition component in the base layer to obtain the user characteristics of the user historical behavior data.

8. The method of claim 1, wherein the recommendation layer comprises a user recommendation layer, an item recommendation layer, and a content recommendation layer; the candidate popularity media data comprises user popularity media data, article popularity media data and content popularity media data;

the identifying the real-time features and the user features based on the recommendation layer, and determining candidate hot media data from the network media data according to the identification result comprises the following steps:

identifying the user characteristics based on the user recommendation layer, scoring the network media data according to a user identification result, and determining the user popularity media data from the network media data according to the user popularity score;

identifying the user characteristics based on the item recommendation layer, carrying out item popularity scoring on the network media data according to an item identification result, and determining the item popularity media data from the network media data according to the item popularity scoring;

Identifying the real-time features based on the content recommendation layer, and performing content labeling processing on the network media data according to a content identification result to obtain a content labeling result;

and determining the content hot media data from the network media data according to the user characteristics and the content labeling result.

9. The method of claim 8, wherein the ordering layer comprises a data deduplication layer and a data ordering layer;

the ranking process for the candidate popularity media data based on the ranking layer comprises the following steps:

performing de-duplication processing on the user heat media data, the article heat media data and the content heat media data based on the data de-duplication layer to obtain target candidate heat media data;

and inputting the target candidate hot media data into the data ordering layer, and ordering the target candidate hot media data based on the data ordering layer.

10. The method of claim 9, wherein the data ordering layers include a forwarding data ordering layer, a comment data ordering layer, and a forward data ordering layer;

the step of inputting the target candidate hot media data into the data sorting layer, and sorting the target candidate hot media data based on the data sorting layer comprises the following steps:

Respectively inputting the target candidate heat media data into the forwarding data sorting layer, the comment data sorting layer and the forward data sorting layer;

determining forwarding probability of the target candidate hot media data based on the forwarding data ordering layer, determining comment probability of the target candidate hot media data based on the comment data ordering layer, and determining forward probability of the target candidate hot media data based on the forward data ordering layer;

weighting and fusing the forwarding probability, the evaluation probability and the forward probability to obtain the fusion probability of the target candidate heat media data;

and sorting the target candidate heat media data based on the fusion probability.

11. The method of claim 1, wherein outputting candidate hotness question-answering items according to the real-time hotness media data comprises:

acquiring heat comment data of the real-time heat media data, and acquiring heat keywords of the heat comment data;

and outputting the candidate hot question and answer items according to the hot keywords.

12. A computer device, comprising: a processor, a memory, and a network interface;

The processor is connected to the memory, the network interface for providing data communication functions, the memory for storing program code, the processor for invoking the program code to perform the steps of the method of any of claims 1 to 11.

13. A computer readable storage medium, characterized in that it stores a computer program comprising program instructions which, when executed by a processor, perform the steps of the method according to any of claims 1 to 11.