Data processing method, device, computer equipment and storage medium

By building account relationship diagrams and graph mining technology, users’ group members on audio and video platforms are automatically identified, solving the problems of high identification costs and poor timeliness in the existing technology, and achieving efficient and automated user group member identification.

CN113139101BActive Publication Date: 2024-11-26TSINGHUA UNIVERSITY +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202110536355.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-05-17
Publication Date
2024-11-26
Estimated Expiration
2041-05-17

AI Technical Summary

Technical Problem

The prior art is difficult to efficiently identify and automatically identify user group members on audio and video platforms, resulting in high cost of manual labeling, poor timeliness and low recall.

Method used

By building an account relationship diagram, seed accounts and target accounts are selected based on interactive behavior, and combined with graph mining technology, the account group members are automatically identified without manual annotation.

Benefits of technology

It greatly saves the labor and time cost of identifying account groups, improves the recognition efficiency, and can automatically update, which has strong timeliness and improves recall rate.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113139101B_ABST
    Figure CN113139101B_ABST
Patent Text Reader

Abstract

The present disclosure relates to a data processing method, device, computer equipment and storage medium, and belongs to the field of computer technology. The present disclosure constructs a corresponding account relationship graph according to the interactive behavior of a first account, and screens out a second account belonging to a certain account group from the first account, and uses the second account and the account relationship graph, combined with graph mining technology, to screen out target accounts belonging to the same account group as the second account from the accounts included in the account relationship graph, without manually marking each account as a group, greatly saving the manpower and time cost of identifying the account group, and improving the efficiency of identifying each account in the account group from a large number of accounts.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to a data processing method, device, computer equipment and storage medium. Background Art

[0002] With the development of computer technology and the diversification of terminal functions, users can use terminals to browse audio and video resources released by creators (such as anchors, up masters, etc.) anytime and anywhere. Some creators may form user groups and be active on audio and video platforms in the form of "Internet celebrity (i.e., Internet celebrity) families", which already have considerable economic influence. Therefore, accurately identifying each user group from the massive registered users of the audio and video platform, and finding each member in the same user group has gradually become a research hotspot.

[0003] Most of the creators in the above user groups are mentors or relatives, and they follow each other and often interact with each other: for example, they appear in each other's video works, initiate joint live broadcasts, and interact in the comment area of ​​works. At present, the creators belonging to the same user group are mainly identified by manual labeling, that is, technical personnel observe the creator's public information and published work information, and combine business experience to determine whether the creator belongs to a certain user group. Summary of the invention

[0004] The present disclosure provides a data processing method, apparatus, computer equipment and storage medium to at least save the labor cost of identifying account groups and improve the efficiency of identifying account groups. The technical solution of the present disclosure is as follows:

[0005] According to one aspect of an embodiment of the present disclosure, there is provided a data processing method, including:

[0006] Based on the interaction behaviors of the plurality of first accounts, an account relationship graph corresponding to the interaction behaviors is obtained, wherein the account relationship graph is used to record accounts that have had the interaction behaviors with the first account, and passively associated accounts of the first account meet the first target condition;

[0007] A plurality of second accounts are obtained by screening from the plurality of first accounts, wherein the second accounts have a mapping relationship with an account group, and the account group is a group consisting of accounts having a target social relationship;

[0008] Based on multiple second accounts and the account relationship graph, at least one target account is screened from the accounts included in the account relationship graph, and the group association parameter of the target account meets the second target condition. The group association parameter is used to characterize the possibility that the target account has a mapping relationship with the account group.

[0009] In a possible implementation manner, the filtering out at least one target account from accounts included in the account relationship graph based on the plurality of second accounts and the account relationship graph includes:

[0010] Based on the plurality of second accounts and the account relationship graph, at least one seed account is obtained from the plurality of second accounts, where the seed account is the second account recorded in the account relationship graph;

[0011] Based on each seed account, at least one third account is screened from the account relationship diagram, and each third account has a corresponding relationship with a seed account;

[0012] The group association parameter of each third account is obtained, and the third account whose group association parameter meets the second target condition is determined as the target account.

[0013] In a possible implementation manner, the selecting at least one seed account from the plurality of second accounts based on the plurality of second accounts and the account relationship graph includes:

[0014] Obtaining an intersection between the account recorded in the account relationship graph and the plurality of second accounts;

[0015] At least one of the seed accounts is obtained by screening out the accounts included in the intersection.

[0016] In a possible implementation manner, the selecting and obtaining at least one seed account from each account included in the intersection includes:

[0017] Based on the number of associated accounts of each account included in the intersection, a weight parameter is assigned to each account, wherein the weight parameter is used to indicate the possibility of each account being screened as a seed account;

[0018] Based on the weight parameters of each account, each account is sampled to obtain at least one seed account.

[0019] In a possible implementation manner, the number of associated accounts includes the number of actively associated accounts, the number of passively associated accounts, and the number of bidirectionally associated accounts, and different types of numbers of associated accounts correspond to different weight ratios;

[0020] The allocating a weight parameter to each account based on the number of associated accounts of each account included in the intersection includes:

[0021] For any account included in the intersection, obtain the number of actively associated accounts, the number of passively associated accounts, and the number of bidirectionally associated accounts of the any account;

[0022] The number of actively associated accounts, the number of passively associated accounts and the number of bidirectionally associated accounts are respectively multiplied by their corresponding weight ratios, and the sum of the multiplied values ​​is determined as the weight parameter of any of the accounts.

[0023] In a possible implementation, in response to any account in the intersection being extracted as a seed account, a weight parameter of the any account is lowered.

[0024] In a possible implementation manner, obtaining the group association parameters of each third account includes:

[0025] For any of the third accounts, based on a degree parameter of the third account in the account relationship graph, obtaining a first association parameter of the third account, wherein the degree parameter is used to indicate the number of edges connected to the third account in the account relationship graph, and the first association parameter is used to indicate the degree of association between the third account and the account group in the account relationship graph;

[0026] Based on the account feature of the third account, obtaining a second association parameter of the third account, where the second association parameter is used to indicate the degree of association between the account feature of the third account and the account group;

[0027] Based on the first association parameter and the second association parameter, a group association parameter of the third account is obtained.

[0028] In a possible implementation, in response to the third account being located in a plurality of the account relationship graphs, acquiring the first association parameter of the third account based on the degree parameter of the third account in the account relationship graph includes:

[0029] Based on the degree distribution of the accounts in each account relationship graph, standardize the degree parameters of the third account in each account relationship graph to obtain multiple standard degree parameters of the third account;

[0030] Based on the relationship graph weights of each account relationship graph, weighting the plurality of standard degree parameters to obtain a plurality of weighted degree parameters;

[0031] Based on the multiple weighted parameters, the first association parameter of the third account is obtained.

[0032] In a possible implementation manner, acquiring the first association parameter of the third account based on the multiple weighted parameters includes:

[0033] Obtaining a sum of the weighted degree parameters;

[0034] In response to the third account belonging to a target account set, determining a value obtained by adding one to the sum as the first association parameter, the target account set being obtained by screening based on text information of the account;

[0035] In response to the third account not belonging to the target account set, the sum value is determined as the first association parameter.

[0036] In a possible implementation manner, the text information of the account includes nickname information and profile information of the account, and the screening process of the target account set includes:

[0037] Acquire at least one first keyword, where the first keyword is used to identify the account group;

[0038] Based on at least one of the first keywords, the target account set that meets the third target condition is obtained from the plurality of the first accounts.

[0039] In a possible implementation manner, the third target condition is that at least one of the nickname information or profile information of the first account contains the first keyword.

[0040] In a possible implementation manner, the second target condition is that the current account is located in the front target position in the order of group association parameters from large to small.

[0041] In a possible implementation manner, after determining the third account whose group association parameter meets the second target condition as the target account, the method further includes:

[0042] Acquire accounts other than the seed account from the plurality of second accounts as a verification account set;

[0043] Obtaining the precision and recall of the target account in the verification account set;

[0044] The steps of screening the seed accounts and obtaining the target accounts are iterated until the precision and recall meet the conditions for stopping the iteration.

[0045] In a possible implementation manner, the screening out a plurality of second accounts from the plurality of first accounts comprises:

[0046] Acquire at least one first keyword and at least one second keyword, wherein the first keyword is used to identify the account group, and the second keyword is used to identify the relationship between different accounts in the account group;

[0047] Based on at least one of the first keywords and at least one of the second keywords, a plurality of the second accounts meeting a fourth target condition are screened from the plurality of the first accounts.

[0048] In a possible implementation manner, the fourth target condition is that both the nickname information and the profile information of the first account contain the first keyword, and the profile information also contains the second keyword.

[0049] According to another aspect of an embodiment of the present disclosure, there is provided a data processing device, including:

[0050] A first acquisition unit is configured to execute interactive behaviors based on multiple first accounts, and acquire an account relationship graph corresponding to the interactive behaviors, wherein the account relationship graph is used to record accounts that have had the interactive behaviors with the first accounts, and passively associated accounts of the first accounts meet the first target condition;

[0051] A first screening unit is configured to screen a plurality of second accounts from the plurality of first accounts, wherein the second accounts have a mapping relationship with an account group, and the account group is a group consisting of accounts having a target social relationship;

[0052] The second screening unit is configured to perform screening based on multiple second accounts and the account relationship diagram, and obtain at least one target account from the accounts included in the account relationship diagram, wherein the group association parameter of the target account meets the second target condition, and the group association parameter is used to characterize the possibility that the target account has a mapping relationship with the account group.

[0053] In a possible implementation manner, the second screening unit includes:

[0054] A first screening subunit is configured to screen at least one seed account from a plurality of second accounts based on the plurality of second accounts and the account relationship graph, wherein the seed account is the second account recorded in the account relationship graph;

[0055] A second screening subunit is configured to screen and obtain at least one third account from the account relationship diagram based on each seed account, each third account having a corresponding relationship with a seed account;

[0056] The acquisition and determination subunit is configured to execute acquisition of group association parameters of each third account, and determine the third account whose group association parameter meets the second target condition as the target account.

[0057] In one possible implementation, the first screening subunit includes:

[0058] A first acquisition sub-sub-unit is configured to execute acquisition of intersections between the account recorded in the account relationship graph and a plurality of the second accounts;

[0059] The screening sub-sub-unit is configured to screen and obtain at least one of the seed accounts from the accounts included in the intersection.

[0060] In one possible implementation, the screening sub-subunit includes:

[0061] an allocating sub-sub-unit, configured to allocate a weight parameter to each account based on the number of associated accounts of each account included in the intersection, wherein the weight parameter is used to indicate the possibility of each account being screened as a seed account;

[0062] The sampling sub-unit is configured to execute sampling of each account based on the weight parameters of each account to obtain at least one seed account.

[0063] In a possible implementation manner, the number of associated accounts includes the number of actively associated accounts, the number of passively associated accounts, and the number of bidirectionally associated accounts, and different types of numbers of associated accounts correspond to different weight ratios;

[0064] The allocation sub-sub-unit is configured to perform:

[0065] For any account included in the intersection, obtain the number of actively associated accounts, the number of passively associated accounts, and the number of bidirectionally associated accounts of the any account;

[0066] The number of actively associated accounts, the number of passively associated accounts and the number of bidirectionally associated accounts are respectively multiplied by their corresponding weight ratios, and the sum of the multiplied values ​​is determined as the weight parameter of any of the accounts.

[0067] In one possible implementation, the allocation sub-sub-unit is further configured to execute:

[0068] In response to any account in the intersection being extracted as a seed account, a weight parameter of the any account is adjusted down.

[0069] In a possible implementation manner, the obtaining and determining subunit includes:

[0070] a second acquisition sub-subunit configured to acquire, for any of the third accounts, a first association parameter of the third account based on a degree parameter of the third account in the account relationship graph, wherein the degree parameter is used to indicate the number of edges connected to the third account in the account relationship graph, and the first association parameter is used to indicate the degree of association between the third account and the account group in the account relationship graph;

[0071] A third acquisition sub-subunit is configured to acquire a second association parameter of the third account based on the account feature of the third account, wherein the second association parameter is used to indicate the degree of association between the account feature of the third account and the account group;

[0072] The fourth acquisition sub-subunit is configured to acquire the group association parameter of the third account based on the first association parameter and the second association parameter.

[0073] In a possible implementation manner, in response to the third account being located in a plurality of the account relationship graphs, the second obtaining sub-subunit includes:

[0074] a standardization sub-sub-unit, configured to perform standardization on the degree parameters of the third account in each account relationship graph based on the degree distribution of the account in each account relationship graph, to obtain a plurality of standard degree parameters of the third account;

[0075] A weighting sub-sub-unit is configured to perform weighting on a plurality of said standard degree parameters based on the relationship graph weights of each account relationship graph to obtain a plurality of weighted degree parameters;

[0076] The obtaining sub-sub-sub-unit is configured to obtain the first association parameter of the third account based on the multiple weighted parameters.

[0077] In one possible implementation, the acquisition sub-sub-unit is configured to execute:

[0078] Obtaining a sum of the weighted degree parameters;

[0079] In response to the third account belonging to a target account set, determining a value obtained by adding one to the sum as the first association parameter, the target account set being obtained by screening based on text information of the account;

[0080] In response to the third account not belonging to the target account set, the sum value is determined as the first association parameter.

[0081] In a possible implementation manner, the text information of the account includes nickname information and profile information of the account, and the screening process of the target account set includes:

[0082] Acquire at least one first keyword, where the first keyword is used to identify the account group;

[0083] Based on at least one of the first keywords, the target account set that meets the third target condition is obtained from the plurality of the first accounts.

[0084] In a possible implementation manner, the third target condition is that at least one of the nickname information or profile information of the first account contains the first keyword.

[0085] In a possible implementation manner, the second target condition is that the current account is located in the front target position in the order of group association parameters from large to small.

[0086] In a possible implementation manner, the device further includes:

[0087] A second acquisition unit is configured to acquire accounts other than the seed account from the plurality of second accounts as a verification account set;

[0088] A third acquisition unit is configured to execute acquisition of the precision and recall of the target account in the verification account set;

[0089] The iteration unit is configured to iteratively perform the steps of screening seed accounts and obtaining target accounts until the precision and recall meet the iteration stop condition.

[0090] In a possible implementation manner, the first screening unit is configured to execute:

[0091] Acquire at least one first keyword and at least one second keyword, wherein the first keyword is used to identify the account group, and the second keyword is used to identify the relationship between different accounts in the account group;

[0092] Based on at least one of the first keywords and at least one of the second keywords, a plurality of the second accounts meeting a fourth target condition are screened from the plurality of the first accounts.

[0093] In a possible implementation manner, the fourth target condition is that both the nickname information and the profile information of the first account contain the first keyword, and the profile information also contains the second keyword.

[0094] According to another aspect of an embodiment of the present disclosure, there is provided a computer device, including:

[0095] one or more processors;

[0096] one or more memories for storing the one or more processor-executable instructions;

[0097] The one or more processors are configured to execute the data processing method in any possible implementation manner of the above aspect.

[0098] According to another aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided. When at least one instruction in the computer-readable storage medium is executed by one or more processors of a computer device, the computer device is enabled to execute the data processing method in any possible implementation of the above-mentioned aspect.

[0099] According to another aspect of an embodiment of the present disclosure, a computer program product is provided, comprising one or more instructions, wherein the one or more instructions can be executed by one or more processors of a computer device, so that the computer device can execute the data processing method in any possible implementation manner of the above-mentioned aspect.

[0100] The technical solution provided by the embodiments of the present disclosure brings at least the following beneficial effects:

[0101] By constructing a corresponding account relationship graph based on the interactive behavior of the first account, and screening out the second account belonging to a certain account group from the first account, using the second account and the account relationship graph, combined with graph mining technology, to screen out the target account that belongs to the same account group as the second account from the accounts included in the account relationship graph, there is no need to manually mark each account as a group, which greatly saves the manpower and time costs of identifying account groups and improves the efficiency of identifying each account in an account group from massive accounts.

[0102] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0103] The drawings herein are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the description are used to explain the principles of the present disclosure, and do not constitute improper limitations on the present disclosure.

[0104] Figure 1 is a schematic diagram of an implementation environment of a data processing method according to an exemplary embodiment;

[0105] Figure 2 is a flow chart of a data processing method according to an exemplary embodiment;

[0106] Figure 3 is a flow chart of a data processing method according to an exemplary embodiment;

[0107] Figure 4 is a flow chart of obtaining target scoring parameters provided by an embodiment of the present disclosure;

[0108] Figure 5 is a logical structure block diagram of a data processing device according to an exemplary embodiment;

[0109] Figure 6 A structural block diagram of a computer device provided by an exemplary embodiment of the present disclosure is shown;

[0110] Figure 7 It is a structural diagram of a computer device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0111] In order to enable ordinary persons in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings.

[0112] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0113] The user information and / or user account information involved in the present disclosure may be information authorized by the user or fully authorized by all parties.

[0114] With the development of computer technology and the diversification of terminal functions, Internet content platforms such as short video platforms and live broadcast platforms have flourished. Online community group mining is a hot issue, that is, how to find closely related smaller user groups in a larger user community. In other words, how to identify individual accounts belonging to the same account group from a large number of platform accounts. Members of the same user group are usually active on the platform in the form of "Internet celebrity families", and most of the members are master-apprentice or kinship relationships. For example, some well-known groups are mainly composed of many mid-level Internet celebrities with tens of thousands to millions of fans, and their leaders are usually top super Vs (certified users with more fans). In a user group, members and leaders are usually called "master and apprentice", and apprentices can also bring apprentices. Members usually indicate their group identity and master-apprentice relationship in their personal introduction pages, and have relationships with other members, and often do activities together: such as appearing in each other's video works, live broadcasting together, and interacting in the comment area of ​​works. User groups represented by internet celebrity families usually have a huge influence on content platforms, can attract huge traffic, and can create very considerable economic benefits.

[0115] Therefore, accurately identifying each member of a user group can help the platform better understand the user community ecology, improve the platform's control over the group, and achieve refined operations. At the same time, the social relationships between different user groups can also reveal the cooperation and interaction between user groups for the platform, and provide information support for downstream businesses such as recommendation, public opinion, and risk control. That is, community mining for user groups has important application value. At present, manual labeling is mainly used to identify user groups. Technical personnel observe the characteristics of users' public information and published works and combine business experience to determine whether a user belongs to a user group. The above manual labeling method requires a large amount of manpower and time costs, and the members of the user group often change, and the manual labeling method is difficult to update in time, so the timeliness is poor. In addition, due to the huge user volume of online communities, the manual labeling method can only cover some members of the user group, resulting in a low overall recall rate.

[0116] In view of this, an embodiment of the present application provides a data processing method, which can mine out target accounts that belong to the same account group as the seed account from a large number of accounts, that is, to mine out other members who belong to the same user group as the core user. This large-scale, automated mining method does not require manual labeling by technical personnel, greatly saves manpower and time costs, and can improve the recall rate of user group members. At the same time, automated community mining can be repeated according to a certain period, and the user group list is continuously updated, which has good timeliness, and is described in detail below.

[0117] Figure 1 is a schematic diagram of an implementation environment of a data processing method according to an exemplary embodiment, see Figure 1 In this implementation environment, a terminal 101 and a server 102 may be included. Both the terminal 101 and the server 102 are computer devices, which will be described in detail below.

[0118] The terminal 101 may be any electronic device that supports multimedia resource services. An application for browsing multimedia resources is installed on the terminal 101. The multimedia resources include but are not limited to audio, video, live broadcast, short video, information, etc. For example, the application may include at least one of a short video application, a live broadcast application, an audio and video application, or a social application. After a user registers an account in the application, he or she may log in to the account in the application and use the account to initiate interactive behaviors, such as publishing his or her own video works, commenting on other people's video works, liking other people's video works, following other people's accounts, giving virtual items in other people's live broadcast rooms, and mentioning (@ function) other people's accounts in comments.

[0119] The terminal 101 and the server 102 may be connected via a wired network or a wireless network.

[0120] The server 102 may be an electronic device for providing background services for the above-mentioned application program, and the server 102 may include at least one of a single server, multiple servers, a cloud computing platform, or a virtualization center. Optionally, the server 102 may undertake the main computing work, and the terminal 101 may undertake the secondary computing work; or, the server 102 undertakes the secondary computing work, and the terminal 101 undertakes the main computing work; or, the terminal 101 and the server 102 use a distributed computing architecture for collaborative computing.

[0121] In some embodiments, the server 102 collects the interactive behaviors of each account corresponding to each terminal, and selects multiple first accounts from a large number of accounts, and constructs multiple account relationship graphs based on the multiple interactive behaviors of the multiple first accounts. Next, the second accounts belonging to a certain account group are selected from the first accounts. Then, for each account relationship graph, a seed account is sampled from the intersection of the account recorded in the account relationship graph and each second account. Finally, based on the graph mining algorithm, the target account belonging to the same account group as the seed account is selected, and the precision and recall are calculated based on the verification account set. The target account mining step is iteratively executed until the precision and recall meet the iteration stop condition or the number of iterations is greater than the number threshold, and the target account mined in the last iteration is output, thereby realizing large-scale, automated mining of each account in the account group (i.e., mining of the user group corresponding to the account group).

[0122] Optionally, the terminal 101 may generally refer to one of a plurality of terminals, and the device type of the terminal 101 includes, but is not limited to, at least one of: a vehicle-mounted terminal, a television, a smart phone, a smart speaker, a tablet computer, an e-book reader, an MP3 (Moving Picture Experts Group Audio Layer III, Moving Picture Experts Group Audio Layer 3) player, an MP4 (Moving Picture Experts Group Audio Layer IV, Moving Picture Experts Group Audio Layer 4) player, a laptop computer, or a desktop computer. In the following embodiments, the terminal 101 includes a smart phone for example.

[0123] Those skilled in the art will appreciate that the number of the terminals 101 may be more or less. For example, there may be only one terminal 101, or there may be dozens or hundreds of terminals 101, or a greater number. The embodiment of the present disclosure does not limit the number and device type of the terminals 101.

[0124] Figure 2 is a flow chart of a data processing method according to an exemplary embodiment. Figure 2The data processing method is executed by a computer device, and is described below using the computer device as a server as an example.

[0125] In step 201, the server obtains an account relationship graph corresponding to the interactive behavior based on the interactive behavior of multiple first accounts, and the account relationship graph is used to record the accounts that have interacted with the first account, and the passively associated accounts of the first account meet the first target condition.

[0126] In step 202, the server screens and obtains a plurality of second accounts from the plurality of first accounts, wherein the second accounts have a mapping relationship with an account group, and the account group is a group consisting of accounts having a target social relationship.

[0127] In step 203, based on the multiple second accounts and the account relationship graph, the server selects at least one target account from the accounts included in the account relationship graph, and the group association parameter of the target account meets the second target condition. The group association parameter is used to characterize the possibility of a mapping relationship between the target account and the account group.

[0128] The method provided by the embodiment of the present disclosure constructs a corresponding account relationship graph according to the interactive behavior of the first account, and screens out a second account belonging to a certain account group from the first account, and uses the second account and the account relationship graph in combination with graph mining technology to screen out target accounts belonging to the same account group as the second account from the accounts included in the account relationship graph. There is no need to manually mark each account as a group, which greatly saves the manpower and time costs of identifying account groups and improves the efficiency of identifying each account in an account group from massive accounts.

[0129] In a possible implementation, based on the plurality of second accounts and the account relationship graph, screening out at least one target account from accounts included in the account relationship graph includes:

[0130] Based on the plurality of second accounts and the account relationship graph, at least one seed account is obtained from the plurality of second accounts, where the seed account is the second account recorded in the account relationship graph;

[0131] Based on each seed account, at least one third account is obtained from the account relationship graph, and each third account has a corresponding relationship with a seed account;

[0132] The group association parameter of each third account is obtained, and the third account whose group association parameter meets the second target condition is determined as the target account.

[0133] In a possible implementation, based on the plurality of second accounts and the account relationship graph, selecting at least one seed account from the plurality of second accounts includes:

[0134] Obtaining the intersection between the account recorded in the account relationship graph and the plurality of second accounts;

[0135] At least one seed account is obtained by screening out the accounts included in the intersection.

[0136] In a possible implementation, at least one seed account is screened from each account included in the intersection and includes:

[0137] Based on the number of associated accounts of each account included in the intersection, a weight parameter is assigned to each account, and the weight parameter is used to indicate the possibility of each account being screened as a seed account;

[0138] Based on the weight parameters of each account, each account is sampled to obtain at least one seed account.

[0139] In a possible implementation manner, the number of associated accounts includes the number of actively associated accounts, the number of passively associated accounts, and the number of bidirectionally associated accounts, and different types of numbers of associated accounts correspond to different weight ratios;

[0140] Based on the number of associated accounts of each account contained in the intersection, weight parameters are assigned to each account including:

[0141] For any account included in the intersection, obtain the number of actively associated accounts, the number of passively associated accounts, and the number of bidirectionally associated accounts of the any account;

[0142] The number of actively associated accounts, the number of passively associated accounts, and the number of bidirectionally associated accounts are multiplied by their corresponding weight ratios, and the sum of the multiplied values ​​is determined as the weight parameter of any account.

[0143] In a possible implementation, in response to any account in the intersection being extracted as a seed account, a weight parameter of the any account is lowered.

[0144] In a possible implementation manner, obtaining the group association parameters of each third account includes:

[0145] For any of the third accounts, based on a degree parameter of the third account in the account relationship graph, obtaining a first association parameter of the third account, wherein the degree parameter is used to indicate the number of edges connected to the third account in the account relationship graph, and the first association parameter is used to indicate the degree of association between the third account and the account group in the account relationship graph;

[0146] Based on the account feature of the third account, obtaining a second association parameter of the third account, where the second association parameter is used to indicate the degree of association between the account feature of the third account and the account group;

[0147] Based on the first association parameter and the second association parameter, a group association parameter of the third account is obtained.

[0148] In a possible implementation, in response to the third account being located in a plurality of account relationship graphs, based on a degree parameter of the third account in the account relationship graph, obtaining a first association parameter of the third account includes:

[0149] Based on the degree distribution of the accounts in each account relationship graph, standardize the degree parameters of the third account in each account relationship graph to obtain multiple standard degree parameters of the third account;

[0150] Based on the relationship graph weights of each account relationship graph, multiple standard degree parameters are weighted to obtain multiple weighted degree parameters;

[0151] Based on the multiple weighted parameters, the first associated parameter of the third account is obtained.

[0152] In a possible implementation, based on the multiple weighted parameters, obtaining the first associated parameter of the third account includes:

[0153] Obtaining the sum of multiple weighted degree parameters;

[0154] In response to the third account belonging to a target account set, determining a value obtained by adding one to the sum as the first association parameter, the target account set being obtained by screening based on text information of the account;

[0155] In response to the third account not belonging to the target account set, the sum value is determined as the first association parameter.

[0156] In a possible implementation, the text information of the account includes nickname information and profile information of the account, and the screening process of the target account set includes:

[0157] Obtaining at least one first keyword, where the first keyword is used to identify the account group;

[0158] Based on at least one of the first keywords, the target account set that meets the third target condition is obtained by screening from the plurality of the first accounts.

[0159] In a possible implementation manner, the third target condition is that at least one of the nickname information or profile information of the first account contains the first keyword.

[0160] In a possible implementation, the second target condition is that the current account is located in the front target position in the order from large to small of the group association parameters.

[0161] In a possible implementation manner, after determining the third account whose group association parameter meets the second target condition as the target account, the method further includes:

[0162] Acquire accounts other than the seed account from the plurality of second accounts as a verification account set;

[0163] Obtain the precision and recall of the target account in the verification account set;

[0164] The steps of screening seed accounts and obtaining target accounts are iterated until the precision and recall meet the conditions for stopping the iteration.

[0165] In a possible implementation manner, screening out a plurality of second accounts from the plurality of first accounts includes:

[0166] Obtain at least one first keyword and at least one second keyword, the first keyword being used to identify the account group, and the second keyword being used to identify the relationship between different accounts in the account group;

[0167] Based on at least one of the first keywords and at least one of the second keywords, a plurality of the second accounts meeting the fourth target condition are screened from the plurality of the first accounts.

[0168] In a possible implementation manner, the fourth target condition is that both the nickname information and the profile information of the first account contain the first keyword, and the profile information also contains the second keyword.

[0169] All the above optional technical solutions can be arbitrarily combined to form optional embodiments of the present disclosure, and will not be described in detail here.

[0170] Figure 3 is a flow chart of a data processing method according to an exemplary embodiment. Figure 3 As shown, the data processing method is executed by a computer device, and is described by taking the computer device as a server as an example. This embodiment includes the following steps.

[0171] In step 301, the server obtains at least one first keyword and at least one second keyword, wherein the first keyword is used for an account group and the second keyword is used to identify the relationship between different accounts in the account group.

[0172] Among them, the account group is a group composed of accounts with target social relationships, and the target social relationship includes but is not limited to at least one of the following: master-apprentice relationship, kinship relationship, partnership relationship, spouse relationship, combination relationship, etc.

[0173] Among them, since the first keyword is used to identify an account group, in other words, the first keyword can also identify the difference between different account groups, so the first keyword is also called a "group keyword". Usually, users with group affiliation will embed the name or code of the group they belong to in the nickname information or profile information of their personal account to declare their own group nature.

[0174] Optionally, the name or code of the group can be a combination of at least one of various characters such as Chinese, English, and numbers. For example, the keywords of XX family can be "XX", "X family", etc. The embodiment of the present disclosure does not specifically limit the form of the noun or code of the group.

[0175] Among them, since the second keyword is used to identify the target social relationship between different accounts in the same account group, the second keyword is also called a "relationship keyword". Usually, users with group affiliation will use relationship keywords to indicate their identity in the user group in the profile information of their personal account.

[0176] Optionally, the target social relationship includes but is not limited to a master-apprentice relationship, a husband-wife relationship, a kinship relationship, a partnership relationship, etc. Taking the master-apprentice relationship as an example, the second keywords corresponding to the master-apprentice relationship include: "master", "master", "sf", etc., and the present disclosed embodiment does not specifically limit this.

[0177] In some embodiments, the server stores a first dictionary and a second dictionary in a database, the first dictionary is used to store at least one of the first keywords, the first dictionary is also called a "group keyword dictionary", and the second dictionary is used to store at least one of the first keywords, the second dictionary is also called a "relationship dictionary". Then, the server reads each first keyword from the first dictionary, and reads each second keyword from the second dictionary.

[0178] In some embodiments, the above-mentioned first dictionary is defined by technical personnel based on business experience, or can also be input by the team leader in the platform information entry interface. The first dictionary records some common names and codes of known user groups. The above-mentioned second dictionary can also be defined by technical personnel based on business experience. The second dictionary records some second keywords with higher frequency of occurrence. The present disclosed embodiment does not make specific limitations on this.

[0179] In the above process, by constructing different dictionary structures to store the first keyword and the second keyword respectively, it is beneficial to efficiently screen out potential second accounts belonging to the account group from the first account, greatly improving the execution efficiency of the algorithm, and facilitating the random addition or deletion of the first keyword or the second keyword, and has good timeliness.

[0180] In some embodiments, the server may not store the first keyword or the second keyword through a dictionary structure, but directly store the character string of the first keyword or the second keyword, which can simplify the data processing flow.

[0181] In step 302, the server screens a target account set that meets a third target condition from a plurality of first accounts based on at least one of the first keywords.

[0182] The target account set is obtained by filtering the text information of the account, and the text information of the account includes the nickname information of the account and the profile information of the account. In some embodiments, the text information of the account may also include the work introduction information of the video work released by the user.

[0183] Optionally, the passively associated account of the first account meets the first target condition. The passively associated account refers to an account that has an associated relationship with the first account without the first account actively triggering the association relationship. For example, the passively associated account refers to an account that follows the first account on the platform (commonly known as a fan account). Optionally, the first target condition means that the number of passively associated accounts is greater than a fan threshold. The fan threshold can be customized by technical personnel (the fan threshold needs to be greater than 0). For example, the fan threshold is 10,000.

[0184] In the above process, by setting the first account as an account with the number of passively associated accounts greater than the threshold number of fans, it is equivalent to preliminarily screening out some first accounts with higher traffic value from the massive accounts in the platform, and then further screening out a target account set that meets the first target condition from the first account based on the first keyword. At this time, the screening efficiency of the target account set can be improved.

[0185] Optionally, the first account may also be all accounts in the platform, which can expand the data capacity and coverage of the target account set and simplify the screening process of the target account set.

[0186] In some embodiments, the third target condition is that at least one of the nickname information or profile information of the first account contains the first keyword. In other words, the nickname information of the first account contains the first keyword, the profile information of the first account contains the first keyword, or the nickname information and profile information of the first account both contain the first keyword. All three situations meet the third target condition.

[0187] Optionally, the server may obtain the nickname information and profile information of each first account, and for each first account, traverse the nickname information and profile information of the first account; if any item of the nickname information or the profile information hits any first keyword of at least one of the first keywords, the current first account is added to the target account set; otherwise, if neither the nickname information nor the profile information hits all the first keywords of at least one of the first keywords, the above detection operation is performed on the nickname information and profile information of the next first account until all the first accounts are traversed.

[0188] Optionally, the server may also obtain the nickname information and profile information of each first account. For each first account, the nickname information of the first account is first traversed. If the nickname information can hit any first keyword in at least one of the first keywords, the current first account is directly added to the target account set; otherwise, if the nickname information does not hit all the first keywords in at least one of the first keywords, the above detection operation is performed on the nickname information of the next first account until all the first accounts are traversed for the first time. Then, the server starts the second traversal, that is, traversing the profile information of the remaining first accounts that have not been added to the target account set. If the profile information can hit any first keyword in at least one of the first keywords, the current first account is added to the target account set; otherwise, if the profile information does not hit all the first keywords in at least one of the first keywords, the above detection operation is performed on the profile information of the next first account until all the first accounts are traversed.

[0189] In some embodiments, the server may also traverse the profile information of all first accounts in the first traversal, and traverse the nickname information of all first accounts in the second traversal, which is not specifically limited in the embodiments of the present disclosure.

[0190] Optionally, the server may also obtain the nickname information and profile information of each first account, and for each first account, first traverse the nickname information of the first account, and if the nickname information has been able to hit any first keyword in at least one of the first keywords, directly add the current first account to the target account set; otherwise, if the nickname information does not hit all the first keywords in at least one of the first keywords, then detect the profile information of the current first account, and if the profile information can hit any first keyword in at least one of the first keywords, still add the current first account to the target account set; otherwise, if the profile information also does not hit all the first keywords in at least one of the first keywords, perform the above detection operation on the nickname information and profile information of the next first account until all the first accounts are traversed.

[0191] Optionally, after obtaining the nickname information and profile information of each first account, the server may also merge the nickname information and profile information of each first account into text information of each first account, and then traverse the text information of each first account. If the text information can hit any first keyword in at least one of the first keywords, the current first account is added to the target account set; otherwise, if the text information does not hit all the first keywords in at least one of the first keywords, the above detection operation is performed on the text information of the next first account until all the first accounts are traversed.

[0192] In the above steps 301-302, the server's screening process for the target account set is shown. Each account in the target account set only needs to contain at least one first keyword in the nickname information or profile information. Therefore, the target account set can also be called a "loose subset" screened out from the first account. The target account set can be used to affect the first associated parameter of each third account in the following step 3091. Please refer to the following step 3091 for details, which will not be repeated here.

[0193] In step 303, the server screens the plurality of first accounts based on at least one of the first keywords and at least one of the second keywords to obtain a plurality of second accounts that meet the fourth target condition.

[0194] Optionally, the first account may be all accounts within the platform, or the first account may also be an account whose number of passively associated accounts is greater than a threshold number of fans, which is not specifically limited in the embodiments of the present disclosure.

[0195] In some embodiments, the fourth target condition is that both the nickname information and the profile information of the first account contain the first keyword, and the profile information also contains the second keyword.

[0196] Optionally, the server may obtain the nickname information and profile information of each first account, and for each first account, traverse the nickname information and profile information of the first account; if the nickname information and the profile information both hit any first keyword of at least one of the first keywords, and the profile information also hits any second keyword of at least one of the second keywords, the current first account is determined to be a second account; otherwise, if any item in the nickname information or the profile information does not hit all the first keywords of at least one of the first keywords, or the profile information does not hit any second keyword of at least one of the second keywords, the above detection operation is performed on the nickname information and profile information of the next first account until all the first accounts are traversed.

[0197] Optionally, after obtaining the nickname information and profile information of each first account, the server may first traverse the nickname information of each first account. If the nickname information can hit any first keyword of at least one of the first keywords, then detect the profile information of the current first account. If the profile information can hit any first keyword of at least one of the first keywords and can also hit any second keyword of at least one of the second keywords, the current first account is determined to be a second account; otherwise, as long as any of the above conditions is not met (for example, the nickname information does not hit any first keyword of at least one of the first keywords, or the profile information does not hit any first keyword of at least one of the first keywords, or the profile information does not hit any second keyword of at least one of the second keywords), the above detection operation is performed on the nickname information and profile information of the next first account until all the first accounts are traversed.

[0198] In the above process, the second account can be screened from the first account through the first keyword and the second keyword. The second account needs to contain at least one first keyword in both the nickname information and the profile information, and the profile information also needs to contain at least one second keyword. Therefore, the set formed by the second account can also be called a "strict subset".

[0199] The above steps 302 and 303 are equivalent to the process of using the first dictionary and the second dictionary to filter the first account to obtain a loose subset and a strict subset respectively. The purpose of dictionary screening is to preliminarily screen out account subsets that are more likely to belong to a certain account group among the numerous first accounts. In the present disclosed embodiment, only the account subset (including a loose subset and a strict subset) of a single account group is taken as an example for explanation. For each account group in multiple account groups, it is only necessary to perform similar screening steps to obtain the account subset of each account group, which will not be repeated here.

[0200] Furthermore, in the following step 304, a strict subset (each second account) can be used to perform snowball sampling on the account relationship network of all users, and a corresponding account relationship graph can be constructed for each interactive behavior to reduce the computational complexity of the algorithm, while the loose subset is used to influence the first association parameter of each third account in the following step 3091 to improve the credibility of the first association parameter. It should be noted that the embodiment of the present disclosure only takes the example of modeling an account relationship graph for an interactive behavior and screening out the target account from this account relationship graph as an example. In actual applications, each first account usually has multiple (i.e., more than one) interactive behaviors. It is only necessary to perform similar operations on each interactive behavior to obtain multiple account relationship graphs, and each target account can be independently screened out for each account relationship graph, which will not be elaborated here.

[0201] The above steps 301-303 are equivalent to the process of screening out multiple second accounts from multiple first accounts, wherein the second accounts have a mapping relationship with the account group, and the account group refers to a designated account group uniquely identified by the first keyword, that is, from a large number of first accounts, for a designated account group, screen out the second accounts that are more likely to belong to the account group.

[0202] In step 304, the server obtains an account relationship graph corresponding to the interactive behavior based on the multiple interactive behaviors of the first account, and the account relationship graph is used to record the accounts that have had the interactive behavior with the first account.

[0203] Optionally, the interactive behavior includes but is not limited to: two-way attention behavior, mutual mention (@ function) behavior, gifting of virtual items in the live broadcast room (for example, the number of gifted virtual items is in the top 10, the amount of gifted virtual items is in the top 10, etc.), mutual comment behavior, mutual like behavior, etc. The disclosed embodiment does not specifically limit the types and number of interactive behaviors.

[0204] In some embodiments, the server obtains multiple interactive behaviors of the first account in a historical time period, for example, the historical time period is from the time of account registration to the present, or the historical time period is the most recent month, etc. The disclosed embodiment does not specifically limit the historical time period. Since some first accounts may have one interactive behavior but not another interactive behavior, the other interactive behavior that has not occurred can be set to NULL or 0, indicating that the second account has not performed the current interactive behavior.

[0205] In some embodiments, for each interactive behavior, the server independently constructs an account relationship network corresponding to the interactive behavior. Each account relationship network includes multiple nodes, each node represents an account, and each node is connected to at least one other node or multiple nodes. The connection between connected nodes is called an edge. Each edge represents that corresponding interactive behaviors have occurred between the accounts represented by the nodes at both ends of the edge. There are no edges between unconnected nodes, that is, no corresponding interactive behaviors have occurred between the accounts represented by the unconnected nodes.

[0206] In an exemplary embodiment, a specified interactive behavior is used as an example for explanation, and the specified interactive behavior can be any interactive behavior generated by the first account. The server uses the multiple second accounts (i.e., the strict subset) selected in the above step 303 to perform snowball sampling on the full amount of account relationship network corresponding to the interactive behavior constructed by the multiple first accounts, so as to sample and obtain the account relationship graph corresponding to the interactive behavior.

[0207] Optionally, the server obtains multiple interactive behaviors of the second account that occurred within the historical time period. In the full account relationship network corresponding to the interactive behavior, for each second account, the node directly connected to the second account is obtained, that is, the account that has had the interactive behavior with the second account is obtained. This process is called the first-level sampling, which is equivalent to sampling the "first-degree friends" of each second account. Then, for each node obtained in the above-mentioned first-level sampling, other nodes directly connected to each node are obtained again. These other nodes are equivalent to being indirectly connected to the second account through the nodes sampled in the first layer. In other words, for the account that has had the interactive behavior with the second account, other accounts that have had the interactive behavior with the account are obtained again. This process is called the second-level sampling, which is equivalent to sampling the "second-degree friends" (that is, friends of friends) of each second account. The above process, that is, using the second account (that is, a strict subset) to perform snowball sampling on the account relationship network of all users, each second account is sampled down two layers, and each interactive behavior is sampled independently, so as to construct the corresponding account relationship graph.

[0208] It should be noted that the above process of constructing the account relationship diagram is only described by taking the example of downsampling 2 layers for the second account. Optionally, the second account may be downsampled 1 layer, 3 layers or any other number of layers, and the present embodiment does not specifically limit this.

[0209] In an exemplary scenario, the server may pre-select the account social relationships of some core users (e.g., users corresponding to the first account), construct a network adjacency table, and construct a separate network adjacency table for each interactive behavior. Then, based on the interactive behaviors recorded in the above network adjacency table, for each interactive behavior of each second account, snowball sampling is performed in the corresponding network adjacency table to obtain the corresponding account relationship graph. Assuming that there are N types of interactive behaviors, the server independently samples each interactive behavior to generate N account relationship graphs: G 1 , G 2 ,……,G N , thus saving the server's computing resources.

[0210] In step 305, the server obtains the intersection between the account recorded in the account relationship graph and the plurality of second accounts.

[0211] In some embodiments, since one or more seed accounts need to be anchored when mining account groups, and the seed account needs to belong to the second account and have appeared in the account relationship graph, this can avoid the situation where the seed account does not have the interaction behavior and the account relationship graph does not include the seed account, which weakens the utilization value of the collected account relationship graph, thereby improving data utilization.

[0212] In some embodiments, the server traverses each node in the account relationship graph, and if any node belongs to the second account, the node is added to the intersection, otherwise the server continues to traverse the next node in the account relationship graph until all nodes in the account relationship graph are traversed to obtain the intersection involved in the above step 305.

[0213] In some embodiments, the server obtains an account node set consisting of all nodes in the account relationship graph, and calculates the intersection between the account node set and the strict subset consisting of each second account to obtain the intersection involved in the above step 305.

[0214] In step 306, the server assigns a weight parameter to each account based on the number of associated accounts of each account included in the intersection, and the weight parameter is used to indicate the possibility of each account being screened as a seed account.

[0215] Among them, the number of associated accounts of an account includes at least one of the following: the number of actively associated accounts (commonly known as one-way attention), the number of passively associated accounts (commonly known as the number of fans), and the number of bidirectionally associated accounts (commonly known as bidirectional attention). Different types of numbers of associated accounts correspond to different weight ratios, and the embodiments of the present disclosure do not specifically limit this.

[0216] Optionally, for any account included in the intersection, the server obtains the number of actively associated accounts, the number of passively associated accounts, and the number of bidirectionally associated accounts of the any account, multiplies the number of actively associated accounts, the number of passively associated accounts, and the number of bidirectionally associated accounts by their respective corresponding weight ratios, and determines the sum of the multiplied values ​​as the weight parameter of the any account.

[0217] In an exemplary embodiment, the server obtains the number of fans, two-way attention and one-way attention of any account, then standardizes the number of fans, two-way attention and one-way attention of any account, and then weightedly sums the number of fans, two-way attention and one-way attention of any account according to their respective corresponding weight ratios to obtain the weight parameter of any account.

[0218] In some embodiments, the above-mentioned weight ratio can be the number of fans: the amount of two-way attention: the amount of one-way attention = 3:2:1. Of course, the above-mentioned weight ratio can also be the number of fans: the amount of two-way attention: the amount of one-way attention = 5:3:1. The embodiments of the present disclosure do not specifically limit the value of the above-mentioned weight ratio.

[0219] In step 307, the server samples each account based on the weight parameter of each account to obtain at least one seed account.

[0220] In some embodiments, the server performs weighted random sampling on each second account according to the weight parameters of each account. Optionally, in the weighted random sampling process, sampling with replacement or sampling without replacement can be adopted, and after the number of seed accounts is set, the number of seed accounts can be selected from the intersection based on the weighted random sampling algorithm.

[0221] For example, the weighted random sampling algorithm is a reservoir algorithm, or the weighted random sampling algorithm can also be an A-Res algorithm, which is not specifically limited in the embodiments of the present disclosure. For example, the number of accounts is any value greater than or equal to 1, for example, the number of accounts is 3.

[0222] In the above process, it is equivalent to taking the intersection of the strict subset and the node set obtained by snowball sampling in the above step 304, and extracting the seed account from the intersection. In some embodiments, the server extracts the same set of seed accounts from the intersection for all account relationship graphs, that is, different account relationship graphs correspond to the same seed account. At this time, it is necessary to ensure that the seed account has appeared in each account relationship graph corresponding to each interactive behavior, or the server independently extracts different groups of seed accounts from the intersection for each account relationship graph, that is, different account relationship graphs correspond to different seed accounts. The embodiments of the present disclosure do not specifically limit this.

[0223] In the above steps 306-307, a possible implementation method is shown in which the server selects at least one seed account from the accounts included in the intersection. In some embodiments, instead of using weighted random sampling, one or more accounts in the intersection may be manually designated as seed accounts to simplify the process of extracting seed accounts. The embodiments of the present disclosure do not specifically limit this.

[0224] In the above process, based on the multiple second accounts and the account relationship graph, the server selects at least one seed account from the multiple second accounts, and the seed account is the second account recorded in the account relationship graph. It should be noted that the above seed account is the key node of the subsequent mining step of the target account in the account group. The seed node where the seed account is located will be used as the center to mine each target account that is closely related to the seed node. Therefore, the seed account must be included in the account relationship graph, and the seed account needs to be the second account at the same time.

[0225] In step 308, the server selects at least one third account from the account relationship graph based on each seed account, and each third account has a corresponding relationship with a seed account.

[0226] In some embodiments, the server may obtain at least one third account corresponding to any seed account from the account relationship graph for any seed account based on the ACL (Anderson-Chung-Lang Algorithm), wherein the ACL algorithm is a local community mining algorithm based on PPR (Personal PageRank).

[0227] Optionally, for any seed account, the server obtains the adjacency matrix of the account relationship graph, and for the account relationship graph, based on the PPR (Personal PageRank, personalized web page ranking) algorithm, obtains the PPR vector of any seed account in the adjacency matrix of the account relationship graph, and the PPR vector is used to measure the degree of connectivity between any seed account and all other account nodes in the account relationship graph, that is, the probability of randomly walking to all other account nodes in the account relationship graph from any seed account. Then, the Sweep Cut algorithm is executed according to the PPR vector of any seed account to filter out the node set (CommunitV) that is closely connected to any seed account, and the above steps are repeated for each seed account until all seed accounts are traversed. Finally, the node sets mined by all seed accounts are combined, and after removing duplicate nodes in the union, the mining result output by the final community mining algorithm (that is, at least one third account) can be obtained.

[0228] In an exemplary embodiment, when there are multiple account relationship graphs, each account relationship graph is mined independently. The above ACL community mining algorithm can be expressed as the following pseudo code:

[0229] Symbols: account group number t, account relationship diagram number i, seed account number j.

[0230] Input: Seed account s of account group t jt (j=1, 2, 3), account relationship graph G i The adjacency matrix A it . Output: Account relationship graph G i Account Group Community it .

[0231] Algorithm starts

[0232] Step 1: Calculate the seed account s jt In the adjacency matrix A it The PPR vector on ijt .

[0233] Step 2: According to PPR ijtThe vector executes the sweep cut algorithm to select the seed account s jt A Community of tightly connected nodes ijt .

[0234] Step 3: Repeat steps 1 and 2 until all seed accounts have been traversed.

[0235] Step 4: Add all seed accounts to the Community ijt After merging and removing duplicate nodes, we get account group t in the account relationship graph G i Community mining results on Community it .

[0236] End of algorithm

[0237] For the analysis of the above algorithm, the embodiment of the present disclosure only takes the process of mining a single account relationship graph for a single given account group as an example to illustrate, and for the same given account group t, if N account relationship graphs are modeled based on N types of interactive behaviors, after mining the N account relationship graphs independently, N community mining results can be obtained, that is, for each account group, at least one third account can be mined on each account relationship graph, and the final output is the union of the N community mining results mined on the N account relationship graphs, that is, Community it (i=1, 2, 3, . . . , N) wherein N is an integer greater than or equal to 1.

[0238] In step 309, the server obtains the group association parameters of each third account, and determines the third account whose group association parameters meet the second target condition as the target account.

[0239] Among them, the group association parameter of the target account meets the second target condition, and the group association parameter is used to characterize the possibility that the target account and the account group have a mapping relationship. In other words, the group association parameter is used to characterize the possibility that the target account and the seed account belong to the same account group.

[0240] The following is a detailed description of the process of obtaining group association parameters.

[0241] Optionally, for any third account among at least one of the third accounts, when the server obtains the group association parameter of any third account, it may first obtain the first association parameter of any third account, then obtain the second association parameter of any third account, and then combine the first association parameter and the second association parameter to obtain the group association parameter of any third account.

[0242] Figure 4This is a flowchart of obtaining group association parameters provided by the embodiment of the present disclosure. Please refer to Figure 4 For any third account among at least one of the third accounts, when the server obtains the group association parameter of the any third account, the following steps 3091-3093 are included.

[0243] In step 3091, the server obtains a first association parameter of any third account based on a degree parameter of the third account in the account relationship graph.

[0244] The degree parameter is used to indicate the number of edges connected to the third account in the account relationship graph.

[0245] The first association parameter is used to indicate the degree of association between the third account and the account group in the account relationship diagram. In other words, the first association parameter is used to characterize the predicted score of whether the third account is the target account based on the account relationship diagram.

[0246] In some embodiments, in response to the third account being located in multiple account relationship graphs, the server may standardize the degree parameters of the third account in each account relationship graph based on the degree distribution of the accounts in each account relationship graph to obtain multiple standard degree parameters of the third account; weight the multiple standard degree parameters based on the relationship graph weights of each account relationship graph to obtain multiple weighted degree parameters; and obtain the first association parameter of the third account based on the multiple weighted degree parameters.

[0247] Optionally, since different nodes in each account relationship graph have different degree parameters, the distribution obeyed by the degree parameters of all nodes is called degree distribution. For each account relationship graph, the degree parameters can be standardized according to the degree distribution to obtain a standard degree parameter. The standardization process is also the regularization (or normalization) process, which converts the standard degree parameter into a value between 0 and 1.

[0248] Optionally, the server may assign a relationship graph weight to each account relationship graph, and the relationship graph weights of different account relationship graphs may be the same or different. For example, the server assigns equal weights to each account relationship graph, or different relationship graph weights may be manually configured by a technician. The embodiments of the present disclosure do not specifically limit this.

[0249] After obtaining the relationship graph weights of each account relationship graph, the standard degree parameters of each third account can be weighted according to the relationship graph weights of each account relationship graph to obtain corresponding weighted degree parameters.

[0250] In some embodiments, the server may determine the sum of the weighted degree parameters as the first association parameter, which can simplify the process of obtaining the first association parameter.

[0251] In some embodiments, the server can also obtain the sum of the multiple weighted parameters; in response to the third account belonging to the target account set, the value obtained by adding one to the sum is determined as the first associated parameter, and the target account set is obtained by filtering the text information of the account; in response to the third account not belonging to the target account set, the sum is determined as the first associated parameter.

[0252] That is to say, on the basis of obtaining the sum of the weighted parameters, if the third account is in the target account set obtained in the above step 302, it is equivalent to adding one to the vote of the third account, giving the third account a larger first association parameter; otherwise, the sum is still used as the first association parameter. In this way, the first association parameter can be refined and adjusted using the loose subset (target account set), thereby improving the accuracy of the first association parameter.

[0253] The above process of obtaining the first association parameter of each third account, that is, the process of performing weighted voting on each third account, for a given account group t, after taking the union of the N communities mined from the N account relationship graphs, the voting rule can be expressed as the following rule:

[0254] 1) Assign a relationship graph weight to each account relationship graph. The default is equal weight, which can also be manually specified. The relationship graph weight is recorded as

[0255] 2) Mining results for each community it The degree distribution of the account in the N community mining results is standardized, and the standardized standard degree parameter of the third account j in the N community mining results is recorded as degree ij , where degree ij ∈[0, 1];

[0256] 3) Use the loose subset (i.e., the target account set) obtained in step 302 to screen all the third accounts. If the third account is in the loose subset, the vote is increased by 1.

[0257] 4) The votes of the third account j are: Wherein, 1 (in relax set) is an indicative function, and when the third account j is in the relaxed subset, 1 (in relax set)=1.

[0258] After voting, each third account gets a score (i.e., votes). After the scores of each third account are standardized, the standardized score is called the first association parameter, also known as the structural score parameter, denoted as Str_Score j .

[0259] In step 3092, the server obtains a second associated parameter of the third account based on the account characteristics of the third account.

[0260] The second association parameter is used to indicate the degree of association between the account characteristics of the third account and the account group. In other words, the second association parameter is used to indicate a predicted score of whether the third account is a target account based on the account characteristics.

[0261] Optionally, the account characteristics of the third account include attribute characteristics of the third account and behavior characteristics of the third account.

[0262] In some embodiments, since the users corresponding to the target accounts in the account group are usually Internet celebrities with relatively fixed content output and personal marketing behaviors, the behavioral characteristics of the target accounts will be different from those of other accounts. Therefore, each third account can be evaluated again from the dimension of account characteristics to evaluate the possibility of each third account being the target account.

[0263] In some embodiments, for any third account, the server obtains the attribute characteristics of the third account: number of fans, one-way attention, two-way attention, etc., and obtains the behavioral characteristics of the third account: the amount of works uploaded in the past week, the live broadcast time in the past week, the amount of gifts in the past week, the number of likes sent in the past week, the number of comments sent in the past week, etc.

[0264] Optionally, the server performs one-hot encoding or embedding encoding on the above attribute features and the above behavior features to obtain the account features of the third account. Optionally, the server stores the account features of all accounts in an account feature library, and then reads the account features of each third account from the account feature library according to the account identifier of each third account.

[0265] Optionally, the server inputs the account features of each third account into a pre-trained logistic regression model, and uses the logistic regression model to predict the probability that "each third account is the target account". This probability is also the second association parameter, which can be called the propensity score parameter, recorded as Prop_Score j Optionally, the above-mentioned logistic regression model is also a binary classification model, which is used to classify the input third account and output the probability that the third account is the target account.

[0266] In step 3093, the server obtains the group association parameter of the third account based on the first association parameter and the second association parameter.

[0267] In some embodiments, the server may obtain the arithmetic mean between the first association parameter and the second association parameter as the group association parameter, or the server may obtain the weighted mean between the first association parameter and the second association parameter as the group association parameter, where the parameter weight is specified by a technician, or the server may obtain the geometric mean between the first association parameter and the second association parameter as the group association parameter, which is not specifically limited in the embodiments of the present disclosure.

[0268] In an exemplary embodiment, taking the group association parameter as an arithmetic mean as an example, the group association parameter Score j It can be expressed as:

[0269]

[0270] After the group association parameters of each third account are obtained through the above steps 3091-3093, the target accounts whose group association parameters meet the second target condition can be screened out based on the group association parameters of each third account.

[0271] Optionally, the second target condition is that the account is in the front target position when the group association parameters are sorted from large to small, or, the second target condition is that the group association parameter is greater than a scoring threshold, and the scoring threshold is any value greater than or equal to 0, or, the second target condition is that the account is in the front target proportion when the group association parameters are sorted from large to small.

[0272] In some embodiments, the server may sort each third account in descending order of the group association parameter, and determine the third account ranked in the front target position as the target account; optionally, the server may also determine the third account whose group association parameter is greater than the scoring threshold as the target account; optionally, the server may also sort each third account in descending order of the group association parameter, and determine the third account ranked in the front target ratio as the target account, which is not specifically limited in the embodiments of the present disclosure. Among them, the target ratio can be defined according to business needs, but it is necessary to ensure that the target ratio is greater than 0 and less than 1.

[0273] In an exemplary embodiment, the server sorts each third account in descending order according to the group association parameter, and defines a target ratio according to business needs to determine the number of target accounts finally output. Optionally, the target ratio is an empirical value of 0.3. At this time, the server outputs the third account ranked in the top 30% as at least one of the target accounts finally screened out.

[0274] In the above steps 304-309, the server screens and obtains at least one target account from the accounts included in the account relationship diagram based on the multiple second accounts and the account relationship diagram. Optionally, the server may directly output at least one target account, or the server may also perform multiple iterations and update the target account by executing the following steps 310-312, and the target account outputted in the final iteration process shall prevail, which is not specifically limited in the embodiments of the present disclosure.

[0275] In step 310, the server obtains accounts other than the seed account from the plurality of second accounts as a verification account set.

[0276] Optionally, the server deletes at least one seed account selected this time from the strict subset formed by the multiple second accounts, and forms a verification account set with the remaining second accounts, and the verification account set is used to detect the recognition effect of the account group.

[0277] In step 311, the server obtains the precision and recall of the target account in the verification account set.

[0278] Among them, the precision rate refers to how many of the predicted target accounts belong to the verification account set. Assuming that the predicted target accounts that belong to the verification account set are called positive classes (TP), and the predicted target accounts that do not belong to the verification account set are called negative classes (FP), then the precision rate P can be expressed as P=TP / (TP+FP).

[0279] Among them, the recall rate refers to how many second accounts in the verification account set are correctly predicted as target accounts. Assuming that the second accounts in the verification account set are predicted as target accounts are called positive accounts (TP), and the second accounts in the verification account set are not predicted as target accounts are called negative accounts (FN), then the recall rate R can be expressed as R=TP / (TP+FN).

[0280] In step 312, the server iteratively performs the steps of screening seed accounts and obtaining target accounts until the precision and recall meet the conditions for stopping iteration.

[0281] Optionally, if the precision and recall rates do not meet the conditions for stopping iteration, the server iteratively executes the above steps 306-311, that is, reselects the seed account and obtains the target account until the precision and recall rates meet the conditions for stopping iteration, or the number of iterations meets the conditions for stopping iteration, and then stops iterating.

[0282] Optionally, the condition for stopping iteration includes that both the precision and the recall are less than 0.1, or the condition for stopping iteration is that the number of iterations exceeds a threshold number, and the threshold number is greater than or equal to 1. The embodiment of the present disclosure does not specifically limit this.

[0283] Optionally, during the iteration process, the server may also update the weight parameters assigned to each account in the intersection in the above step 306, and the update logic includes: in response to any account in the intersection being extracted as a seed account, lowering the weight parameter of any account, for example, modifying the weight parameter of any account to half of the original weight parameter.

[0284] In other words, the update logic of the weight parameters can be expressed as the following formula:

[0285]

[0286] Among them, w j ′ is the updated weight parameter, w j is the original weight parameter.

[0287] In some embodiments, when either the precision or the recall is greater than or equal to 0.1, the iteration is stopped and the mining result of the last iteration is output as the final mining result, or when the number of iterations exceeds a threshold, the iteration is stopped. Since the ideal result has not been mined when the iteration is stopped, the second accounts in the strict subset can be directly sorted according to the second association parameter output by the logistic regression model, and the second accounts ranked at the front target proportion can be output as the final mining result.

[0288] In some embodiments, the above iterative process may have exceptions, mainly including dictionary screening exceptions and network sampling exceptions, which are analyzed below.

[0289] 1) Dictionary screening failure: When the number of second accounts in the strict subset is less than 3, the seed account cannot be screened out. In order to ensure that the automatic iterative check can be executed normally, the number of second accounts in the strict subset is required to be at least 5. If the number of second accounts in the strict subset is less than 5, the algorithm will report an exception. The reason for the exception may be that the dictionary is expired or the setting is unreasonable. The error prompt message can be set to "The number of second accounts in the strict subset is less than 3, please check the dictionary", and skip the current account group and mine the next account group.

[0290] 2) Network sampling anomaly: If the number of accounts in the intersection of the strict subset and the accounts included in the current account relationship graph is less than 3, the seed account cannot be selected from it, the algorithm reports an exception, and outputs an error message "The number of candidate seed accounts is less than 3, the current account relationship graph is too small, and mining fails", and skips the current account relationship graph, and mines the account group on the next account relationship graph.

[0291] All the above optional technical solutions can be arbitrarily combined to form optional embodiments of the present disclosure, and will not be described in detail here.

[0292] The method provided by the embodiment of the present disclosure constructs a corresponding account relationship graph according to the interactive behavior of the first account, and screens out a second account belonging to a certain account group from the first account, and uses the second account and the account relationship graph in combination with graph mining technology to screen out target accounts belonging to the same account group as the second account from the accounts included in the account relationship graph. There is no need to manually mark each account as a group, which greatly saves the manpower and time costs of identifying account groups and improves the efficiency of identifying each account in an account group from massive accounts.

[0293] In the above process, the data processing method can effectively mine the account groups of the platform. Compared with the manually labeled data set, the algorithm recognition accuracy rate reaches 82% and the recall rate reaches 135%. The recall rate exceeds 100% because the manually labeled data set used in the experiment missed the mark of the account group members, and this method was used to find the missing group members. It can be seen that this method can greatly reduce the workload of manual labeling and more comprehensively reflect the member structure of the account group. Using this method, the user group information corresponding to the mined account group can be used in business areas such as recommendation, search, and public opinion. It also provides basic information support for the operation and business analysis of user groups and Internet celebrity families.

[0294] Figure 5 is a logical structure block diagram of a data processing device according to an exemplary embodiment. Figure 5 , the device comprises:

[0295] The first acquisition unit 501 is configured to execute an interactive behavior based on multiple first accounts, and acquire an account relationship graph corresponding to the interactive behavior, wherein the account relationship graph is used to record accounts that have had the interactive behavior with the first account, and the passively associated accounts of the first account meet the first target condition;

[0296] The first screening unit 502 is configured to screen and obtain a plurality of second accounts from the plurality of first accounts, wherein the second accounts have a mapping relationship with an account group, and the account group is a group consisting of accounts having a target social relationship;

[0297] The second screening unit 503 is configured to perform screening based on the multiple second accounts and the account relationship diagram, and obtain at least one target account from the accounts included in the account relationship diagram, wherein the group association parameter of the target account meets the second target condition, and the group association parameter is used to characterize the possibility that the target account has a mapping relationship with the account group.

[0298] The device provided by the embodiment of the present disclosure constructs a corresponding account relationship graph according to the interactive behavior of the first account, and screens out a second account belonging to a certain account group from the first account, and uses the second account and the account relationship graph in combination with graph mining technology to screen out target accounts belonging to the same account group as the second account from the accounts included in the account relationship graph. There is no need to manually mark each account as a group, which greatly saves the manpower and time costs of identifying account groups and improves the efficiency of identifying each account in an account group from massive accounts.

[0299] In one possible implementation, based on Figure 5 The second screening unit 503 comprises:

[0300] A first screening subunit is configured to screen the plurality of second accounts based on the plurality of second accounts and the account relationship graph to obtain at least one seed account, where the seed account is the second account recorded in the account relationship graph;

[0301] A second screening subunit is configured to perform screening based on each seed account to obtain at least one third account from the account relationship diagram, each third account having a corresponding relationship with a seed account;

[0302] The acquisition and determination subunit is configured to execute acquisition of group association parameters of each third account, and determine the third account whose group association parameter meets the second target condition as the target account.

[0303] In one possible implementation, based on Figure 5 The first screening subunit comprises:

[0304] A first acquisition sub-sub-unit is configured to execute acquisition of intersections between the account recorded in the account relationship graph and the plurality of second accounts;

[0305] The screening sub-sub-unit is configured to screen and obtain at least one of the seed accounts from the accounts included in the intersection.

[0306] In one possible implementation, based on Figure 5 The device is composed of the screening sub-subunit comprising:

[0307] an allocating sub-sub-unit, configured to allocate a weight parameter to each account based on the number of associated accounts of each account included in the intersection, wherein the weight parameter is used to indicate the possibility of each account being screened as a seed account;

[0308] The sampling sub-unit is configured to execute a weight parameter based on each account, sample each account, and obtain at least one seed account.

[0309] In a possible implementation manner, the number of associated accounts includes the number of actively associated accounts, the number of passively associated accounts, and the number of bidirectionally associated accounts, and different types of numbers of associated accounts correspond to different weight ratios;

[0310] The allocation sub-sub-unit is configured to perform:

[0311] For any account included in the intersection, obtain the number of actively associated accounts, the number of passively associated accounts, and the number of bidirectionally associated accounts of the any account;

[0312] The number of actively associated accounts, the number of passively associated accounts, and the number of bidirectionally associated accounts are multiplied by their corresponding weight ratios, and the sum of the multiplied values ​​is determined as the weight parameter of any account.

[0313] In one possible implementation, the allocation sub-sub-unit is further configured to execute:

[0314] In response to any account in the intersection being extracted as a seed account, a weight parameter of the any account is adjusted lower.

[0315] In one possible implementation, based on Figure 5 The device is composed of, the acquisition determination subunit includes:

[0316] A second acquisition sub-subunit is configured to execute, for any of the third accounts, based on a degree parameter of the third account in the account relationship graph, to acquire a first association parameter of the third account, wherein the degree parameter is used to indicate the number of edges connected to the third account in the account relationship graph, and the first association parameter is used to indicate the degree of association between the third account and the account group in the account relationship graph;

[0317] A third acquisition sub-subunit is configured to acquire a second association parameter of the third account based on the account feature of the third account, where the second association parameter is used to indicate the degree of association between the account feature of the third account and the account group;

[0318] The fourth acquisition sub-subunit is configured to acquire the group association parameter of the third account based on the first association parameter and the second association parameter.

[0319] In one possible implementation, in response to the third account being located in a plurality of account relationship graphs, based on Figure 5 The second acquisition sub-subunit comprises:

[0320] a standardization sub-sub-unit, configured to perform standardization of degree parameters of the third account in each account relationship graph based on the degree distribution of the accounts in each account relationship graph, to obtain a plurality of standard degree parameters of the third account;

[0321] A weighting sub-sub-unit is configured to perform a weighting of a relationship graph based on each account relationship graph, weighting a plurality of the standard degree parameters to obtain a plurality of weighted degree parameters;

[0322] The obtaining sub-sub-sub-unit is configured to obtain the first associated parameter of the third account based on the multiple weighted parameters.

[0323] In one possible implementation, the acquisition sub-sub-unit is configured to execute:

[0324] Obtaining the sum of multiple weighted degree parameters;

[0325] In response to the third account belonging to a target account set, determining a value obtained by adding one to the sum as the first association parameter, the target account set being obtained by screening based on text information of the account;

[0326] In response to the third account not belonging to the target account set, the sum value is determined as the first association parameter.

[0327] In a possible implementation, the text information of the account includes nickname information and profile information of the account, and the screening process of the target account set includes:

[0328] Obtaining at least one first keyword, where the first keyword is used to identify the account group;

[0329] Based on at least one of the first keywords, the target account set that meets the third target condition is obtained by screening from the plurality of the first accounts.

[0330] In a possible implementation manner, the third target condition is that at least one of the nickname information or profile information of the first account contains the first keyword.

[0331] In a possible implementation, the second target condition is that the current account is located in the front target position in the order from large to small of the group association parameters.

[0332] In one possible implementation, based on Figure 5 The device is composed of:

[0333] A second acquisition unit is configured to acquire accounts other than the seed account from the plurality of second accounts as a verification account set;

[0334] A third acquisition unit is configured to execute acquisition of the precision and recall of the target account in the verification account set;

[0335] The iteration unit is configured to iteratively perform the steps of screening seed accounts and obtaining target accounts until the precision and recall meet the conditions for stopping iteration.

[0336] In a possible implementation, the first screening unit 502 is configured to execute:

[0337] Obtain at least one first keyword and at least one second keyword, the first keyword being used to identify the account group, and the second keyword being used to identify the relationship between different accounts in the account group;

[0338] Based on at least one of the first keywords and at least one of the second keywords, a plurality of the second accounts meeting the fourth target condition are screened from the plurality of the first accounts.

[0339] In a possible implementation manner, the fourth target condition is that both the nickname information and the profile information of the first account contain the first keyword, and the profile information also contains the second keyword.

[0340] Regarding the device in the above embodiment, the specific manner in which each unit performs the operation has been described in detail in the embodiment of the data processing method, and will not be elaborated here.

[0341] Figure 6 FIG. 1 shows a block diagram of a computer device provided by an exemplary embodiment of the present disclosure. Figure 6 As shown, a computer device is taken as a terminal 600 for example. The terminal 600 may be a smart phone, a tablet computer, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 player (Moving Picture Experts Group Audio Layer IV), a laptop computer or a desktop computer. The terminal 600 may also be called a user device, a portable terminal, a laptop terminal, a desktop terminal or other names.

[0342] Typically, the terminal 600 includes a processor 601 and a memory 602 .

[0343] The processor 601 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 601 may be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 601 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a CPU (Central Processing Unit); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor 601 may be integrated with a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 601 may also include an AI (Artificial Intelligence) processor, which is used to process computing operations related to machine learning.

[0344] The memory 602 may include one or more computer-readable storage media, which may be non-transitory. The memory 602 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In some embodiments, the non-transitory computer-readable storage medium in the memory 602 is used to store at least one instruction, which is used to be executed by the processor 601 to implement the data processing method provided in each embodiment of the present disclosure.

[0345] In some embodiments, the terminal 600 may further optionally include: a peripheral device interface 603 and at least one peripheral device. The processor 601, the memory 602 and the peripheral device interface 603 may be connected via a bus or a signal line. Each peripheral device may be connected to the peripheral device interface 603 via a bus, a signal line or a circuit board. Specifically, the peripheral device includes: at least one of a radio frequency circuit 604, a touch display screen 605, a camera assembly 606, an audio circuit 607, a positioning assembly 608 and a power supply 609.

[0346] The peripheral device interface 603 may be used to connect at least one peripheral device related to I / O (Input / Output) to the processor 601 and the memory 602. In some embodiments, the processor 601, the memory 602, and the peripheral device interface 603 are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor 601, the memory 602, and the peripheral device interface 603 may be implemented on a separate chip or circuit board, which is not limited in this embodiment.

[0347] The radio frequency circuit 604 is used to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The radio frequency circuit 604 communicates with the communication network and other communication devices through electromagnetic signals. The radio frequency circuit 604 converts the electrical signal into an electromagnetic signal for transmission, or converts the received electromagnetic signal into an electrical signal. Optionally, the radio frequency circuit 604 includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a user identity module card, and the like. The radio frequency circuit 604 can communicate with other terminals through at least one wireless communication protocol. The wireless communication protocol includes, but is not limited to: a metropolitan area network, various generations of mobile communication networks (2G, 3G, 4G and 5G), a wireless local area network and / or a WiFi (Wireless Fidelity) network. In some embodiments, the radio frequency circuit 604 may also include circuits related to NFC (Near Field Communication), which is not limited in the present disclosure.

[0348] The display screen 605 is used to display the UI (User Interface). The UI may include graphics, text, icons, videos and any combination thereof. When the display screen 605 is a touch display screen, the display screen 605 also has the ability to collect touch signals on the surface or above the surface of the display screen 605. The touch signal can be input to the processor 601 as a control signal for processing. At this time, the display screen 605 can also be used to provide virtual buttons and / or virtual keyboards, also known as soft buttons and / or soft keyboards. In some embodiments, the display screen 605 can be one, and the front panel of the terminal 600 is set; in other embodiments, the display screen 605 can be at least two, which are respectively set on different surfaces of the terminal 600 or are folded; in some other embodiments, the display screen 605 can be a flexible display screen, which is set on the curved surface or folded surface of the terminal 600. Even, the display screen 605 can also be set to a non-rectangular irregular shape, that is, a special-shaped screen. The display screen 605 can be made of materials such as LCD (Liquid Crystal Display), OLED (Organic Light-Emitting Diode, organic light-emitting diode).

[0349] The camera assembly 606 is used to capture images or videos. Optionally, the camera assembly 606 includes a front camera and a rear camera. Typically, the front camera is arranged on the front panel of the terminal, and the rear camera is arranged on the back of the terminal. In some embodiments, there are at least two rear cameras, which are any one of a main camera, a depth of field camera, a wide-angle camera, and a telephoto camera, so as to realize the fusion of the main camera and the depth of field camera to realize the background blur function, the fusion of the main camera and the wide-angle camera to realize the panoramic shooting and VR (Virtual Reality) shooting function or other fusion shooting functions. In some embodiments, the camera assembly 606 may also include a flash. The flash can be a monochrome temperature flash or a dual-color temperature flash. A dual-color temperature flash refers to a combination of a warm light flash and a cold light flash, which can be used for light compensation at different color temperatures.

[0350] The audio circuit 607 may include a microphone and a speaker. The microphone is used to collect sound waves from the user and the environment, and convert the sound waves into electrical signals and input them into the processor 601 for processing, or input them into the radio frequency circuit 604 to achieve voice communication. For the purpose of stereo acquisition or noise reduction, there may be multiple microphones, which are respectively arranged at different parts of the terminal 600. The microphone may also be an array microphone or an omnidirectional acquisition microphone. The speaker is used to convert the electrical signal from the processor 601 or the radio frequency circuit 604 into sound waves. The speaker may be a traditional film speaker or a piezoelectric ceramic speaker. When the speaker is a piezoelectric ceramic speaker, it can not only convert the electrical signal into sound waves audible to humans, but also convert the electrical signal into sound waves inaudible to humans for purposes such as ranging. In some embodiments, the audio circuit 607 may also include a headphone jack.

[0351] Positioning component 608 is used to locate the current geographical location of terminal 600 to implement navigation or LBS (Location Based Service). Positioning component 608 can be a positioning component based on the US GPS (Global Positioning System), China's Beidou system, Russia's Grenas system or the European Union's Galileo system.

[0352] The power supply 609 is used to power various components in the terminal 600. The power supply 609 can be an alternating current, a direct current, a disposable battery, or a rechargeable battery. When the power supply 609 includes a rechargeable battery, the rechargeable battery can support wired charging or wireless charging. The rechargeable battery can also be used to support fast charging technology.

[0353] In some embodiments, the terminal 600 further includes one or more sensors 610 , including but not limited to: an acceleration sensor 611 , a gyroscope sensor 612 , a pressure sensor 613 , a fingerprint sensor 614 , an optical sensor 615 , and a proximity sensor 616 .

[0354] The acceleration sensor 611 can detect the magnitude of acceleration on the three coordinate axes of the coordinate system established by the terminal 600. For example, the acceleration sensor 611 can be used to detect the components of gravity acceleration on the three coordinate axes. The processor 601 can control the touch display screen 605 to display the user interface in a horizontal view or a vertical view according to the gravity acceleration signal collected by the acceleration sensor 611. The acceleration sensor 611 can also be used for collecting game or user motion data.

[0355] The gyro sensor 612 can detect the body direction and rotation angle of the terminal 600, and the gyro sensor 612 can cooperate with the acceleration sensor 611 to collect the user's 3D actions on the terminal 600. The processor 601 can implement the following functions based on the data collected by the gyro sensor 612: motion sensing (such as changing the UI according to the user's tilt operation), image stabilization during shooting, game control, and inertial navigation.

[0356] The pressure sensor 613 can be set on the side frame of the terminal 600 and / or the lower layer of the touch display screen 605. When the pressure sensor 613 is set on the side frame of the terminal 600, it can detect the user's holding signal of the terminal 600, and the processor 601 performs left and right hand recognition or shortcut operation according to the holding signal collected by the pressure sensor 613. When the pressure sensor 613 is set on the lower layer of the touch display screen 605, the processor 601 controls the operability controls on the UI interface according to the user's pressure operation on the touch display screen 605. The operability controls include at least one of a button control, a scroll bar control, an icon control, and a menu control.

[0357] The fingerprint sensor 614 is used to collect the user's fingerprint, and the processor 601 identifies the user's identity based on the fingerprint collected by the fingerprint sensor 614, or the fingerprint sensor 614 identifies the user's identity based on the collected fingerprint. When the user's identity is identified as a trusted identity, the processor 601 authorizes the user to perform relevant sensitive operations, including unlocking the screen, viewing encrypted information, downloading software, paying, and changing settings. The fingerprint sensor 614 can be set on the front, back, or side of the terminal 600. When a physical button or a manufacturer logo is set on the terminal 600, the fingerprint sensor 614 can be integrated with the physical button or the manufacturer logo.

[0358] The optical sensor 615 is used to collect the ambient light intensity. In one embodiment, the processor 601 can control the display brightness of the touch display screen 605 according to the ambient light intensity collected by the optical sensor 615. Specifically, when the ambient light intensity is high, the display brightness of the touch display screen 605 is increased; when the ambient light intensity is low, the display brightness of the touch display screen 605 is reduced. In another embodiment, the processor 601 can also dynamically adjust the shooting parameters of the camera assembly 606 according to the ambient light intensity collected by the optical sensor 615.

[0359] The proximity sensor 616, also called a distance sensor, is usually arranged on the front panel of the terminal 600. The proximity sensor 616 is used to collect the distance between the user and the front of the terminal 600. In one embodiment, when the proximity sensor 616 detects that the distance between the user and the front of the terminal 600 is gradually decreasing, the processor 601 controls the touch display screen 605 to switch from the screen-on state to the screen-off state; when the proximity sensor 616 detects that the distance between the user and the front of the terminal 600 is gradually increasing, the processor 601 controls the touch display screen 605 to switch from the screen-off state to the screen-on state.

[0360] Those skilled in the art will understand that Figure 6 The structure shown in the figure does not constitute a limitation on the terminal 600, and the terminal 600 may include more or less components than those shown in the figure, or combine some components, or adopt a different component arrangement.

[0361] Figure 7 It is a structural diagram of a computer device provided by an embodiment of the present disclosure. The computer device 700 may have relatively large differences due to different configurations or performances, and may include one or more processors (Central Processing Units, CPU) 701 and one or more memories 702, wherein the memory 702 stores at least one program code, and the at least one program code is loaded and executed by the processor 701 to implement the data processing methods provided by the above-mentioned various embodiments. Of course, the computer device 700 may also have components such as a wired or wireless network interface, a keyboard, and an input and output interface for input and output. The computer device 700 may also include other components for implementing device functions, which will not be repeated here.

[0362] In an exemplary embodiment, a computer-readable storage medium including at least one instruction is also provided, such as a memory including at least one instruction, and the at least one instruction can be executed by a processor in a computer device to complete the data processing method in the above embodiment. Optionally, the computer-readable storage medium can be a non-temporary computer-readable storage medium, for example, the non-temporary computer-readable storage medium can include ROM (Read-Only Memory), RAM (Random-Access Memory), CD-ROM (Compact Disc Read-Only Memory), magnetic tape, floppy disk and optical data storage device, etc.

[0363] In an exemplary embodiment, a computer program product is also provided, including one or more instructions, which can be executed by a processor of a computer device to complete the data processing methods provided by the above embodiments.

[0364] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are to be considered exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

[0365] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A data processing method, It is characterized in that include: Based on the interaction behaviors of the plurality of first accounts, an account relationship graph corresponding to the interaction behaviors is obtained, wherein the account relationship graph is used to record accounts that have had the interaction behaviors with the first account, and passively associated accounts of the first account meet the first target condition; A plurality of second accounts are obtained by screening from the plurality of first accounts, wherein the second accounts have a mapping relationship with an account group, and the account group is a group consisting of accounts having a target social relationship; Based on the plurality of second accounts and the account relationship graph, at least one seed account is obtained from the plurality of second accounts, where the seed account is the second account recorded in the account relationship graph; Based on each seed account, at least one third account is screened from the account relationship diagram, and each third account has a corresponding relationship with a seed account; Obtain group association parameters of each third account, and determine the third account whose group association parameters meet the second target condition as the target account, wherein the group association parameters are used to characterize the possibility that the target account has a mapping relationship with the account group, and the second target condition is that the current account is in the front target position when the group association parameters are sorted from large to small.

2. The method according to claim 1, It is characterized in that The selecting at least one seed account from the plurality of second accounts based on the plurality of second accounts and the account relationship graph comprises: Obtaining an intersection between the account recorded in the account relationship graph and the plurality of second accounts; At least one of the seed accounts is obtained by screening out the accounts included in the intersection.

3. The method according to claim 2, It is characterized in that The step of selecting at least one seed account from each account included in the intersection includes: Based on the number of associated accounts of each account included in the intersection, a weight parameter is assigned to each account, wherein the weight parameter is used to indicate the possibility of each account being screened as a seed account; Based on the weight parameters of each account, each account is sampled to obtain at least one seed account.

4. The method according to claim 3, It is characterized in that The number of associated accounts includes the number of actively associated accounts, the number of passively associated accounts, and the number of bidirectionally associated accounts, and different types of associated account numbers correspond to different weight ratios; The allocating a weight parameter to each account based on the number of associated accounts of each account included in the intersection includes: For any account included in the intersection, obtain the number of actively associated accounts, the number of passively associated accounts, and the number of bidirectionally associated accounts of the any account; The number of actively associated accounts, the number of passively associated accounts and the number of bidirectionally associated accounts are respectively multiplied by their corresponding weight ratios, and the sum of the multiplied values ​​is determined as the weight parameter of any of the accounts.

5. The method according to claim 3 or 4, It is characterized in that In response to any account in the intersection being extracted as a seed account, a weight parameter of the any account is adjusted down.

6. The method according to claim 1, It is characterized in that The step of obtaining group association parameters of each third account includes: For any of the third accounts, based on a degree parameter of the third account in the account relationship graph, obtaining a first association parameter of the third account, wherein the degree parameter is used to indicate the number of edges connected to the third account in the account relationship graph, and the first association parameter is used to indicate the degree of association between the third account and the account group in the account relationship graph; Based on the account feature of the third account, obtaining a second association parameter of the third account, where the second association parameter is used to indicate the degree of association between the account feature of the third account and the account group; Based on the first association parameter and the second association parameter, a group association parameter of the third account is obtained.

7. The method according to claim 6, It is characterized in that In response to the third account being located in the plurality of account relationship graphs, the acquiring the first associated parameter of the third account based on the degree parameter of the third account in the account relationship graphs includes: Based on the degree distribution of the accounts in each account relationship graph, standardize the degree parameters of the third account in each account relationship graph to obtain multiple standard degree parameters of the third account; Based on the relationship graph weights of each account relationship graph, weighting the plurality of standard degree parameters to obtain a plurality of weighted degree parameters; Based on the multiple weighted parameters, the first association parameter of the third account is obtained.

8. The method according to claim 7, It is characterized in that The acquiring the first associated parameter of the third account based on the plurality of weighted parameters comprises: Obtaining a sum of the weighted degree parameters; In response to the third account belonging to a target account set, determining a value obtained by adding one to the sum as the first association parameter, the target account set being obtained by screening based on text information of the account; In response to the third account not belonging to the target account set, the sum value is determined as the first association parameter.

9. The method according to claim 8, It is characterized in that The text information of the account includes the nickname information and the profile information of the account. The screening process of the target account set includes: Acquire at least one first keyword, where the first keyword is used to identify the account group; Based on at least one of the first keywords, the target account set that meets the third target condition is obtained by screening from the plurality of the first accounts.

10. The method according to claim 9, It is characterized in that The third target condition is that at least one of the nickname information or profile information of the first account contains the first keyword.

11. The method according to claim 1, It is characterized in that After determining the third account whose group association parameter meets the second target condition as the target account, the method further includes: Acquire accounts other than the seed account from the plurality of second accounts as a verification account set; Obtaining the precision and recall of the target account in the verification account set; The steps of screening the seed accounts and obtaining the target accounts are iterated until the precision and recall meet the conditions for stopping the iteration.

12. The method according to claim 1, It is characterized in that The screening of the plurality of second accounts from the plurality of first accounts comprises: Acquire at least one first keyword and at least one second keyword, wherein the first keyword is used to identify the account group, and the second keyword is used to identify the relationship between different accounts in the account group; Based on at least one of the first keywords and at least one of the second keywords, a plurality of the second accounts meeting a fourth target condition are screened from the plurality of the first accounts.

13. The method according to claim 12, It is characterized in that The fourth target condition is that both the nickname information and the profile information of the first account contain the first keyword, and the profile information also contains the second keyword.

14. A data processing device, It is characterized in that include: A first acquisition unit is configured to execute interactive behaviors based on multiple first accounts, and acquire an account relationship graph corresponding to the interactive behaviors, wherein the account relationship graph is used to record accounts that have had the interactive behaviors with the first accounts, and passively associated accounts of the first accounts meet the first target condition; A first screening unit is configured to screen a plurality of second accounts from the plurality of first accounts, wherein the second accounts have a mapping relationship with an account group, and the account group is a group consisting of accounts having a target social relationship; A second screening unit, comprising a first screening sub-unit, a second screening sub-unit and an acquisition determination sub-unit; The first screening subunit is configured to screen at least one seed account from the plurality of second accounts based on the plurality of second accounts and the account relationship graph, wherein the seed account is the second account recorded in the account relationship graph; The second screening subunit is configured to screen and obtain at least one third account from the account relationship diagram based on each seed account, each third account having a corresponding relationship with a seed account; The acquisition and determination subunit is configured to execute acquisition of group association parameters of each third account, and determine the third account whose group association parameters meet the second target condition as the target account, the group association parameters are used to characterize the possibility that the target account has a mapping relationship with the account group, and the second target condition is that the current account is in the front target position when the group association parameters are sorted from large to small.

15. The device according to claim 14, It is characterized in that The first screening subunit comprises: A first acquisition sub-sub-unit is configured to execute acquisition of intersections between the account recorded in the account relationship graph and a plurality of the second accounts; The screening sub-sub-unit is configured to screen and obtain at least one of the seed accounts from the accounts included in the intersection.

16. The device according to claim 15, It is characterized in that The screening sub-subunit comprises: an allocating sub-sub-unit, configured to allocate a weight parameter to each account based on the number of associated accounts of each account included in the intersection, wherein the weight parameter is used to indicate the possibility of each account being screened as a seed account; The sampling sub-unit is configured to execute sampling of each account based on the weight parameters of each account to obtain at least one seed account.

17. The device according to claim 16, It is characterized in that The number of associated accounts includes the number of actively associated accounts, the number of passively associated accounts, and the number of bidirectionally associated accounts, and different types of associated account numbers correspond to different weight ratios; The allocation sub-sub-unit is configured to perform: For any account included in the intersection, obtain the number of actively associated accounts, the number of passively associated accounts, and the number of bidirectionally associated accounts of the any account; The number of actively associated accounts, the number of passively associated accounts and the number of bidirectionally associated accounts are respectively multiplied by their corresponding weight ratios, and the sum of the multiplied values ​​is determined as the weight parameter of any of the accounts.

18. The device according to claim 16 or 17, It is characterized in that The allocation sub-sub-unit is further configured to perform: In response to any account in the intersection being extracted as a seed account, a weight parameter of the any account is adjusted down.

19. The device according to claim 14, It is characterized in that The acquisition determination subunit comprises: a second acquisition sub-subunit configured to acquire, for any of the third accounts, a first association parameter of the third account based on a degree parameter of the third account in the account relationship graph, wherein the degree parameter is used to indicate the number of edges connected to the third account in the account relationship graph, and the first association parameter is used to indicate the degree of association between the third account and the account group in the account relationship graph; A third acquisition sub-subunit is configured to acquire a second association parameter of the third account based on the account feature of the third account, wherein the second association parameter is used to indicate the degree of association between the account feature of the third account and the account group; The fourth acquisition sub-subunit is configured to acquire the group association parameter of the third account based on the first association parameter and the second association parameter.

20. The device according to claim 19, It is characterized in that In response to the third account being located in the plurality of account relationship graphs, the second obtaining sub-subunit includes: a standardization sub-sub-unit, configured to perform standardization on the degree parameters of the third account in each account relationship graph based on the degree distribution of the account in each account relationship graph, to obtain a plurality of standard degree parameters of the third account; A weighting sub-sub-unit is configured to perform weighting on a plurality of said standard degree parameters based on the relationship graph weights of each account relationship graph to obtain a plurality of weighted degree parameters; The obtaining sub-sub-sub-unit is configured to obtain the first association parameter of the third account based on the multiple weighted parameters.

21. The device according to claim 20, It is characterized in that The Get sub-sub-sub-unit is configured to perform: Obtaining a sum of the weighted degree parameters; In response to the third account belonging to a target account set, determining a value obtained by adding one to the sum as the first association parameter, the target account set being obtained by screening based on text information of the account; In response to the third account not belonging to the target account set, the sum value is determined as the first association parameter.

22. The device according to claim 21, It is characterized in that The text information of the account includes the nickname information and the profile information of the account. The screening process of the target account set includes: Acquire at least one first keyword, where the first keyword is used to identify the account group; Based on at least one of the first keywords, the target account set that meets the third target condition is obtained by screening from the plurality of the first accounts.

23. The device according to claim 22, It is characterized in that The third target condition is that at least one of the nickname information or profile information of the first account contains the first keyword.

24. The device according to claim 14, It is characterized in that The device also includes: A second acquisition unit is configured to acquire accounts other than the seed account from the plurality of second accounts as a verification account set; A third acquisition unit is configured to execute acquisition of the precision and recall of the target account in the verification account set; The iteration unit is configured to iteratively perform the steps of screening seed accounts and obtaining target accounts until the precision and recall meet the iteration stop condition.

25. The device according to claim 14, It is characterized in that The first screening unit is configured to perform: Acquire at least one first keyword and at least one second keyword, wherein the first keyword is used to identify the account group, and the second keyword is used to identify the relationship between different accounts in the account group; Based on at least one of the first keywords and at least one of the second keywords, a plurality of the second accounts meeting a fourth target condition are screened from the plurality of the first accounts.

26. The device according to claim 25, It is characterized in that The fourth target condition is that both the nickname information and the profile information of the first account contain the first keyword, and the profile information also contains the second keyword.

27. A computer device, It is characterized in that include: one or more processors; one or more memories for storing the one or more processor-executable instructions; The one or more processors are configured to execute the instructions to implement the data processing method according to any one of claims 1 to 13.

28. A computer readable storage medium, It is characterized in that When at least one instruction in the computer-readable storage medium is executed by one or more processors of a computer device, the computer device is enabled to execute the data processing method according to any one of claims 1 to 13.

29. A computer program product, It is characterized in that The method comprises one or more instructions, wherein the one or more instructions are executed by one or more processors of a computer device, so that the computer device can execute the data processing method according to any one of claims 1 to 13.