CN112616074B - Target group identification method and electronic equipment - Google Patents

Target group identification method and electronic equipment Download PDF

Info

Publication number
CN112616074B
CN112616074B CN202110248766.1A CN202110248766A CN112616074B CN 112616074 B CN112616074 B CN 112616074B CN 202110248766 A CN202110248766 A CN 202110248766A CN 112616074 B CN112616074 B CN 112616074B
Authority
CN
China
Prior art keywords
node
path
account
nodes
target
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202110248766.1A
Other languages
Chinese (zh)
Other versions
CN112616074A (en
Inventor
王璐
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wuhan Douyu Network Technology Co Ltd
Original Assignee
Wuhan Douyu Network Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Wuhan Douyu Network Technology Co Ltd filed Critical Wuhan Douyu Network Technology Co Ltd
Priority to CN202110248766.1A priority Critical patent/CN112616074B/en
Publication of CN112616074A publication Critical patent/CN112616074A/en
Application granted granted Critical
Publication of CN112616074B publication Critical patent/CN112616074B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/25—Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
    • H04N21/258—Client or end-user data management, e.g. managing client capabilities, user preferences or demographics, processing of multiple end-users preferences to derive collaborative data
    • H04N21/25866—Management of end-user data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q20/00—Payment architectures, schemes or protocols
    • G06Q20/08—Payment architectures
    • G06Q20/10—Payment architectures specially adapted for electronic funds transfer [EFT] systems; specially adapted for home banking systems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q20/00—Payment architectures, schemes or protocols
    • G06Q20/30—Payment architectures, schemes or protocols characterised by the use of specific devices or networks
    • G06Q20/36—Payment architectures, schemes or protocols characterised by the use of specific devices or networks using electronic wallets or electronic money safes
    • G06Q20/367—Payment architectures, schemes or protocols characterised by the use of specific devices or networks using electronic wallets or electronic money safes involving electronic purses or money safes
    • G06Q20/3678—Payment architectures, schemes or protocols characterised by the use of specific devices or networks using electronic wallets or electronic money safes involving electronic purses or money safes e-cash details, e.g. blinded, divisible or detecting double spending
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/4508—Management of client data or end-user data
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
    • H04N21/4662—Learning process for intelligent management, e.g. learning user preferences for recommending movies characterized by learning algorithms
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
    • H04N21/4667—Processing of monitored end-user data, e.g. trend analysis based on the log file of viewer selections
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/47—End-user applications
    • H04N21/475—End-user interface for inputting end-user data, e.g. personal identification number [PIN], preference data
    • H04N21/4751—End-user interface for inputting end-user data, e.g. personal identification number [PIN], preference data for defining user accounts, e.g. accounts for children
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/47—End-user applications
    • H04N21/478—Supplemental services, e.g. displaying phone caller identification, shopping application

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Databases & Information Systems (AREA)
  • Accounting & Taxation (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Finance (AREA)
  • General Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Child & Adolescent Psychology (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Human Computer Interaction (AREA)
  • Computer Graphics (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The embodiment of the invention discloses a target group identification method and electronic equipment, wherein the method comprises the following steps: determining a circular path for transferring the target prop between account number nodes according to the behavior of presenting the target prop mutually among users; determining an aggregation metric value of each account node in the annular path based on the similarity of the set characteristic dimensions among the account nodes in the annular path; and determining whether a user group corresponding to each account node in the annular path is a target group according to the aggregative metric value. By the technical scheme of the embodiment of the invention, the target user group is accurately identified and searched, and the technical effect of identification cost is reduced.

Description

Target group identification method and electronic equipment
Technical Field
The embodiment of the invention relates to the technical field of live broadcast rooms, in particular to a target group identification method and electronic equipment.
Background
On a platform, such as a live platform, there may be some activity to transfer funds using virtual currency. Usually, the act is to recharge the illegal funds into virtual money and then transfer the virtual money to the own live broadcast room through a plurality of gift giving acts, so that the illegal funds are washed out after a plurality of transfers.
The above behaviors are extremely harmful to the platform, and need to be identified timely and accurately, and corresponding measures are taken to maintain the rights and interests of the platform.
Disclosure of Invention
The embodiment of the invention provides a target group identification method and electronic equipment, which are used for accurately identifying a target group with illegal fund transfer behaviors.
In a first aspect, an embodiment of the present invention provides a target group identification method, where the method includes:
determining a circular path for transferring the target prop between account number nodes according to the behavior of presenting the target prop mutually among users;
determining an aggregation metric value of each account node in the annular path based on the similarity of the set characteristic dimensions among the account nodes in the annular path;
and determining whether a user group corresponding to each account node in the annular path is a target group according to the aggregative metric value.
In a second aspect, an embodiment of the present invention further provides an apparatus for identifying a target group, where the apparatus includes:
the annular path determining module is used for determining an annular path for transferring the target prop between the account number nodes according to the behavior of mutually presenting the target prop between users;
the aggregation metric value determining module is used for determining the aggregation metric value of each account node in the annular path based on the similarity of the set characteristic dimension between the account nodes in the annular path;
and the target group determining module is used for determining whether a user group corresponding to each account node in the annular path is a target group according to the aggregative metric value.
In a third aspect, an embodiment of the present invention further provides an electronic device, where the electronic device includes:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement a target community identification method according to any one of the embodiments of the present invention.
In a fourth aspect, the embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the target community identification method according to any one of the embodiments of the present invention.
According to the technical scheme of the embodiment of the invention, the annular path of the target prop transferred between the account number nodes is determined according to the behavior of the target prop given by the users mutually, the aggregation metric value of each account number node in the annular path is determined based on the similarity of the set characteristic dimension between the account number nodes in the annular path, and then whether the user group corresponding to each account number node in the annular path is the target group is determined according to the aggregation metric value, so that the problems of high complexity and poor real-time performance when the target group is determined by constructing a complex graph relation are solved, the target group is accurately identified and searched, and the technical effect of identification cost is reduced.
Drawings
In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, a brief description is given below of the drawings used in describing the embodiments. It should be clear that the described figures are only views of some of the embodiments of the invention to be described, not all, and that for a person skilled in the art, other figures can be derived from these figures without inventive effort.
Fig. 1 is a flowchart illustrating a target group identification method according to an embodiment of the present invention;
fig. 2 is a flowchart illustrating a target group identification method according to a second embodiment of the present invention;
FIG. 3 is a schematic diagram of a forward map construction provided by the second embodiment of the present invention;
FIG. 4 is a diagram illustrating a circular path search according to a second embodiment of the present invention;
fig. 5 is a schematic structural diagram of a target group identification apparatus according to a third embodiment of the present invention;
fig. 6 is a schematic structural diagram of an electronic device according to a fourth embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Example one
Fig. 1 is a flowchart of a target group identification method according to an embodiment of the present invention, which is applicable to a case of identifying a group with an improper fund transfer behavior in a platform, where the platform may be a live broadcast platform or other platform including a reward or gift operation, and the method may be executed by a target group identification apparatus, which may be implemented by software and/or hardware, and is generally integrated in a terminal, for example, a server corresponding to the platform.
As shown in fig. 1, the method specifically includes the following steps:
s110, determining a circular path for transferring the target prop between account number nodes according to the mutual giving behavior of the target prop between users.
The user can be a user of the platform, and the platform can be a live broadcast platform, a novel reading platform and other platforms containing operations such as appreciation, gift delivery and the like. The target prop may be a prop that circulates as virtual funds within the platform, such as: the fins in the platform are directly sown by the goby. The account number may be an account number in the platform, may include a user account number of the platform, and may further include a live broadcast account number or an author account number registered by each user. The circular path may be a closed loop formed by transferring the target prop between account nodes, and it may be understood that in the circular path, the target prop passes through the same node at least twice in the transferring process.
Specifically, the time, the number, the source, the destination and other information of mutual presentation of the props among the users can be determined according to the prop transfer record data logs. The transfer condition of the target prop between users can be determined according to the information in the recorded data, a transfer path diagram of the target prop can be further constructed according to the transfer condition of the target prop, and a circular path is determined in the transfer path diagram and used for subsequently identifying the use of a group with illegal fund transfer behaviors.
S120, determining the aggregation metric value of each account node in the annular path based on the similarity of the set characteristic dimensions among the account nodes in the annular path.
The set feature dimension may be a preset feature used for identifying a target group, specifically, a feature related to target prop transfer, and may include a target prop quantity acquired by each account node within a preset time, a target prop quantity transferred by each account node within a preset time, and the like. The aggregative metric value can be used for measuring the similarity degree of each account node on the annular path and the target community account node on the set characteristic dimension.
Specifically, according to a predetermined set feature dimension, the similarity between the account nodes in the circular path in the set feature dimension may be determined in a manner of cosine similarity, pearson correlation coefficient, euclidean distance, or the like. After determining the similarity of the set characteristic dimensions between account number nodes in the circular path, the aggregation metric of each account number node in the circular path may be determined according to the average value or weighted average value of the similarity of each account number node in each characteristic dimension, so as to measure the characteristic similarity of each account number node in the circular path. The larger the aggregative metric value is, the larger the similarity of the set characteristic dimensions of the account nodes on the circular path is, and the higher the possibility that the account nodes in the circular path belong to a group is.
The aggregative metric value of each account node in the ring path may be determined based on the following formula:
Figure 635266DEST_PATH_IMAGE001
wherein the content of the first and second substances,
Figure 3668DEST_PATH_IMAGE002
is an aggregative metric value of each account node in the ring path,
Figure 285745DEST_PATH_IMAGE003
is the account on the annular pathNumber nodeuAndvin the first placeiThe similarity in the dimension of each set feature,
Figure 314881DEST_PATH_IMAGE004
is the firstiThe similarity weight of the feature dimension is set,
Figure 555369DEST_PATH_IMAGE005
is the firstiA feature vector corresponding to each set feature dimension,cis a set of account nodes in the circular path,Iis a set of characteristics that are characteristic of the image,
Figure 178112DEST_PATH_IMAGE006
is the total number of account nodes in the ring path. For a set feature dimension of a numerical class,
Figure 857355DEST_PATH_IMAGE007
(ii) a For a set feature dimension of a binary class,
Figure 616363DEST_PATH_IMAGE008
。
specifically, adopt
Figure 90070DEST_PATH_IMAGE009
Calculating the sum of the similarity of each account node in the annular path, specifically the similarity of each account node on the set characteristic dimension
Figure 167485DEST_PATH_IMAGE010
And obtaining the weight.
Figure 119261DEST_PATH_IMAGE011
Is the firstiThe similarity weight of each set feature dimension may represent the importance of the set feature dimension, and the more important set feature dimension corresponds to a higher similarity weight. After the sum of the similarity of each account node in the annular path is obtained through calculation, the aggregation metric value of each account node in the annular path can be determined by dividing the sum of the similarity by the number of the account node pairs in the annular path. According to the ring shapeTotal number of account nodes in path
Figure 998355DEST_PATH_IMAGE012
The number of account node pairs in the ring path may be determined as
Figure 642963DEST_PATH_IMAGE013
。
Illustratively, the total number of account nodes in the circular path is 3, the set characteristic dimension is 2, and the similarity of any two account nodes in the circular path in each set characteristic dimensionFShould have 3, the first setting characteristic dimension corresponds to
Figure 974718DEST_PATH_IMAGE014
Respectively as follows: 0.3, 0.2 and 0.4, with a similarity weight of 3.6; with a second set of feature dimensions
Figure 933447DEST_PATH_IMAGE015
Respectively as follows: 0, 1 and 0, and the similarity weight is 2.5. At this time, the aggregative metric value of each account node in the ring path is
Figure 791681DEST_PATH_IMAGE016
For better understanding of the similarity weight of each set feature dimension in the technical solution of the present embodiment, the similarity weight of each set feature dimension may be determined based on the following formula:
Figure 482557DEST_PATH_IMAGE017
wherein the content of the first and second substances,
Figure 596881DEST_PATH_IMAGE018
is the firstiThe similarity weight of the feature dimension is set,
Figure 93722DEST_PATH_IMAGE019
is the firstiSet characteristic dimensionThe feature vector corresponding to the degree is calculated,
Figure 478567DEST_PATH_IMAGE020
is that
Figure 668239DEST_PATH_IMAGE005
The transposed vector of (a) is,
Figure 99221DEST_PATH_IMAGE021
is a first matrix of the degree of similarity,
Figure 337435DEST_PATH_IMAGE022
is a second similarity matrix of the first and second images,
Figure 904683DEST_PATH_IMAGE021
and
Figure 265257DEST_PATH_IMAGE022
can be determined according to the following formula
Figure 58901DEST_PATH_IMAGE023
Figure 859061DEST_PATH_IMAGE024
Wherein the content of the first and second substances,
Figure 749657DEST_PATH_IMAGE025
represents the first in the first similarity matrixjGo to the firstkThe value of the column is such that,
Figure 77870DEST_PATH_IMAGE026
representing the second in the second similarity matrixjGo to the firstkThe value of the column is such that,llabels representing the account nodes in the determined target community,nwhich represents the number of account number nodes,
Figure 93230DEST_PATH_IMAGE027
indicating having a labellThe number of account number nodes of (1),
Figure 63460DEST_PATH_IMAGE028
is shown asjThe label of each account number node is used,
Figure 808562DEST_PATH_IMAGE029
is shown askLabels of individual account nodes.
Specifically, the denominator in the formula for determining the similarity weight of the set characteristic dimension represents the variance of the account node in the target group and the account node in the non-target group in the set characteristic dimension, and the larger the variance is, the higher the distinctiveness is between the account node in the target group and the account node in the non-target group in the set characteristic dimension. The numerator in the formula for determining the similarity weight of the set characteristic dimension represents the variance of the account number nodes in the target group in each set characteristic dimension, and the smaller the variance is, the higher the similarity of the account number nodes in the target group in the set characteristic dimension is. Therefore, the larger the value of the similarity weight of the set feature dimension is, the greater the degree of distinction of the set feature dimension between the target group and the non-target group is, the greater the importance of the set feature dimension is.
It should be noted that, in the following description,
Figure 183043DEST_PATH_IMAGE030
the similarity matrix is a first similarity matrix, and may specifically be an intra-class similarity matrix, which may be used to represent the similarity of each account node in the target group.
Figure 75913DEST_PATH_IMAGE031
The second similarity matrix may be specifically an inter-class similarity matrix, and may be used to represent the similarity between each account node in the target group and each account node in the non-target group. The label of each account node in the target group with verified history or determined by the business strong rule can be set aslSetting labels of the rest account number nodes as label removallOther labels than, e.g. labelsmAccount sections for labeling non-target groupsAnd (4) point. The first similarity matrix and the second similarity matrix are fixed and unchangeable when calculating the similarity weights of different set characteristic dimensions, namely the first similarity matrix and the second similarity matrix are not changed along with the change of the set characteristic dimensions.
Illustratively, the tag matrix of each account node in the live broadcast platform is [ 2 ]m，m，m，l，m，l]Then it can be determined that there is a taglNumber of account nodes
Figure 928462DEST_PATH_IMAGE027
And 2, the number of account nodes of the live broadcast platform is 6. Further, a first similarity matrix may be determined
Figure 324809DEST_PATH_IMAGE030
Is composed of
Figure 634305DEST_PATH_IMAGE032
A second similarity matrix may be determined
Figure 748891DEST_PATH_IMAGE031
Is composed of
Figure 733028DEST_PATH_IMAGE033
Further, can be based on
Figure 124826DEST_PATH_IMAGE021
And
Figure 231322DEST_PATH_IMAGE031
and determining similarity weight of the set characteristic dimension.
S130, determining whether a user group corresponding to each account node in the annular path is a target group according to the aggregative metric value.
Wherein the target group may be a group in which an illicit funds-transfer act exists.
Specifically, after determining the aggregative metric value of each account node in the circular path, the user group corresponding to each account node in the circular path may be identified according to a preset aggregative metric value threshold. If the aggregative metric value is greater than or equal to a preset aggregative metric value threshold, a user group corresponding to each account node in the annular path can be used as a target group; if the aggregative metric value is smaller than the preset aggregative metric value threshold, the user group corresponding to each account node in the annular path can be used as a non-target group.
It should be noted that, the higher the aggregative metric threshold is set, the more likely some target groups are not identified; the lower the aggregative metric threshold is set, the more likely it is that a normal user will be identified as a target community user. The aggregative metric value threshold may be set according to the requirement of the platform or the requirement of the superior department, and is not particularly limited in this embodiment.
On the basis of the technical solutions of the above embodiments, operation restriction may be performed on each user in the target group, for example: and (3) performing consumption freezing on each account number and a live broadcast room in the target group, and performing operations such as recharging, withdrawing money, sending the target prop and/or receiving the target prop.
According to the technical scheme of the embodiment of the invention, the annular path of the target prop transferred between the account number nodes is determined according to the behavior of the target prop given by the users mutually, the aggregation metric value of each account number node in the annular path is determined based on the similarity of the set characteristic dimension between the account number nodes in the annular path, and then whether the user group corresponding to each account number node in the annular path is the target group is determined according to the aggregation metric value, so that the problems of high complexity and poor real-time performance when the target group is determined by constructing a complex graph relation are solved, the target group is accurately identified and searched, and the technical effect of identification cost is reduced.
Example two
Fig. 2 is a flowchart illustrating a target group identification method according to a second embodiment of the present invention, and reference may be made to the technical solution of this embodiment for a specific implementation manner of determining a circular path for a target prop to transfer between account nodes based on the above embodiment. Wherein explanations of the same or corresponding terms as those of the above-described embodiments are omitted.
As shown in fig. 2, the method specifically includes the following steps:
s210, according to the mutual giving-away behavior of the target props among the users, determining a primary account node initiating the giving-away behavior and a given terminal account node.
The primary account node may be an account node that sends the target item, that is, an account node that initiates a donation action, and the terminal account node may be an account node that receives the target item, that is, an account node that is donated.
Specifically, the time, the number, the source, the destination and other information of mutual presentation of the props among the users can be determined according to the prop transfer record data. And determining a primary account node and a terminal account node in the information of the recorded data.
And S220, according to the initial account number node and the terminal account number node, executing a depth-first search (DFS) algorithm based on the stored forward graph to obtain a ring-shaped path consisting of the initial account number node, at least one intermediate account number node and the terminal account number node.
The intermediate account node can be an account node through which the target prop is transferred from the terminal account node to the initial account node, the forward graph is constructed based on transfer data of the target prop among the account nodes in a set historical time period and is used for determining the transfer condition of the target prop in the historical time period, and the forward graph can be dynamically updated according to the real-time transfer condition of the target prop.
Specifically, after the initial account node and the terminal account node are determined, a path for transferring the target prop from the initial account node to the terminal account node can be determined. A pre-stored forward graph is called, a path of a target prop transferred from a terminal account node to a primary account node through at least one intermediate account node can be determined based on a Depth-First-Search (DFS) algorithm in the forward graph, and the determined two paths are spliced to obtain a circular path formed by the primary account node, the at least one intermediate account node and the terminal account node.
For example, when the initial account node is a and the terminal account node is E, it may be determined that the target prop is transferred from a to E, that is, a > E. According to the forward diagram, the transfer path of the target prop from the terminal account node E to the primary account node A can be determined to be E & gt D & gt C & gt B & gt A, and the account node D, C and B are intermediate account nodes. The two paths can form a ring path A > E > D > C > B > A.
Optionally, the forward graph may be constructed according to the following manner based on transfer data of the target prop between account nodes in a set historical time period:
step one, determining that account nodes of target item presentation and presented behaviors exist in a set historical time period, and taking the account nodes as original nodes of a forward graph.
The set history time period is a time period corresponding to required data for constructing the forward graph, and may be, for example, one day. The specific setting of the set history period may be set according to the real-time performance and accuracy required for identifying the target group, and the longer the set history period is, the higher the accuracy of identifying the target group is, but the more complicated the graph relation page in the forward graph is, the larger the real-time calculation amount is in identifying the target group, and the worse the real-time performance is.
Specifically, the account node where the target item presentation and presented behaviors exist in the set historical time period is used as an original node of the forward graph to construct the forward graph for use. For example: the user account number for presenting the target item can be used as an original node of the forward graph, and the account number of the live broadcast room for receiving the presented target item can be used as the original node.
And step two, aiming at each original node, if the target prop is transferred from the first original node to the second original node, generating an edge pointing to the second original node from the first original node between the first original node and the second original node, and thus constructing the forward graph.
The first original node may be an original node corresponding to an account node of a donor of the target prop, and the second original node may be an original node corresponding to an account node of a receiver of the target prop.
Specifically, according to the transfer of the target prop between the original nodes, an edge corresponding to the transfer behavior of the target prop can be generated. According to the transfer behaviors of all target props in the set historical time period, edges among all original nodes can be generated, and further, all forward graphs generated by the edges are used as constructed forward graphs.
Illustratively, account nodes with target item presentation and presented behaviors in a historical time period are set as a user account A and a live broadcast room account B, the user account A is used as a primary node A of the forward graph, and the live broadcast room account B is used as a primary node B of the forward graph. If the user account a presents the target property to the live broadcast room account B, an edge pointing from the original node a to the original node B may be generated. If the user account a creates a live broadcast room account B, an edge pointing from the original node a to the original node B and an edge pointing from the original node B to the original node a may be generated.
Illustratively, as shown in fig. 3, a node 2 is a user account node, a user corresponding to the node 2 creates a live broadcast account node 1, a node 3 is a live broadcast account node, the user account node 2 donates a target item to the live broadcast account node 3, an owner of the live broadcast account node 3 is a user account node 4, and the user account node 4 donates the target item to the live broadcast account node 1 through the live broadcast account node 3. By analyzing fig. 3, it can be determined that the transfer process of the target prop is 1 > 2 > 3 > 4 > 1, and a circular path is formed.
Specifically, a user account presenting the target item and a presented live broadcast room account can be determined according to the acquired data log transferred by the target item, so that an edge between an original node corresponding to the user account and an original node corresponding to the live broadcast room account is generated. According to the record of the user account in the creation of the live broadcast room, two edges between the original node corresponding to the user account and the original node corresponding to the account in the live broadcast room can be generated.
In order to avoid the time consumed for generating the forward graph according to the account nodes with the target item presenting and presented behaviors in the set historical time period, the forward graph can be dynamically updated according to the set information.
Wherein the setting information includes at least one of: real-time information transfer of the target prop between account number nodes; logout information of the account node; the effective time of the edge between the original nodes.
Specifically, the updating process of the forward map is specifically described according to different setting information. And if the setting information comprises real-time transfer information of the target prop between the account number nodes, generating edges corresponding to the real-time transfer information in the forward graph according to the real-time transfer information. If the setting information includes the logout information of the account number node, the original node and the edge corresponding to the logout account number node in the forward graph can be deleted. If the setting information includes the effective time length of the edges between the original nodes, the life cycle of each edge can be determined, and when the existence time length of a certain edge exceeds the effective time length, the edge is deleted. The action time corresponding to each edge is recorded, the action time is summed with the preset effective time length and is compared with the current time, and if the sum of the action time and the effective time length is greater than the current time, the edge corresponding to the action is deleted. The effective duration may be set as a constant, and may be related to the real-time performance and coverage rate of target community identification, and the specific setting of the effective duration is not specifically limited in this embodiment. The higher the real-time requirement is, the less the stored graph relation is, and the shorter the effective time can be; the higher the coverage requirement, the more graph relationships are stored, and the greater the validity period may be.
To describe the determination of the circular path more clearly, the following steps may be implemented:
step one, taking the terminal account number node as an initial point, executing a DFS algorithm based on a forward graph to search a path, stopping the execution of the DFS algorithm when a first set condition is met, and recording the searched first path.
The first setting condition is a preset condition for stopping the path search, for example, stopping the search when a certain account node is searched.
Specifically, the terminal account node is used as an initial point, a path search is performed in a pre-stored forward graph based on a DFS algorithm from the initial point, when the search result meets a first set condition, the search based on the DFS algorithm is stopped, the path from the initial point to the time when the first set condition is met is recorded, and the path is used as a first path for determining a ring path in the follow-up process.
And step two, with the initial account number node as an initial point, executing the DFS algorithm to search for a path based on the reverse graph of the forward graph, stopping the execution of the DFS algorithm when a second set condition is met, and recording the searched second path.
The second setting condition is a preset condition for stopping the path search, for example, stopping the search when a certain account node is searched.
Specifically, the initial account node is used as an origin, a path search is performed in a pre-stored reverse graph of the forward graph based on the DFS algorithm from the origin, when the search result meets a second set condition, the search based on the DFS algorithm is stopped, the path from the origin to the time when the second set condition is met is recorded, and the path is used as a second path for determining a circular path in the following.
And step three, determining a ring-shaped path according to the first path and the second path.
Specifically, a connection path between the initial account node and the terminal account node is spliced with a reverse path of the recorded first path and second path, and a ring path including the initial account node and the terminal account node is determined.
For more clearly and accurately determining the circular path, the first setting condition may include at least one of: the initial account node is traversed to the hot point node or traversed to the hot point node; the second set condition includes at least one of: the terminal account node is traversed to or to the hotspot node.
If the first set condition is that the initial account node is traversed, it is indicated that a first path from the terminal account node to the initial account node is searched, and the first path and a connection path of the initial account node and the terminal account node are spliced, so that a ring-shaped path including the initial account node and the terminal account node can be determined. If the second set condition is that the terminal account node is traversed, it is indicated that a second path from the initial account node to the terminal account node is searched, and the reverse path of the second path is spliced with the connection path of the initial account node and the terminal account node, so that an annular path including the initial account node and the terminal account node can be determined.
If the first set condition and the second set condition include traversing to a hot node, determining a third path between the first hot node in the first path and a second hot node in the second path based on a hot node path management library; and splicing the reverse path of the first path and the second path and the third path to obtain an annular path.
When the heat value of a certain original node reaches a preset heat threshold value, the original node is used as a hot point node, and paths between the hot point node and other original nodes are more. The hot point node path management library may be a repository storing paths between hot point nodes, and may be updated according to changes of the forward graph.
Specifically, if the first setting condition and the second setting condition that are met are both traversed to the hot spot node, the hot spot node corresponding to the first path may be taken as the first hot spot node, and the hot spot node corresponding to the second path may be taken as the second hot spot node. And finding a path which takes the first hot spot node as an initial point and the second hot spot node as an end point in the hot spot node path pipeline library, and taking the path as a third path. And then splicing the reverse paths of the first path and the second path and the third path to obtain a ring-shaped path.
Illustratively, as shown in fig. 4, the primary account node is 4, and the terminal account node is 1. Taking the terminal account number node 1 as an initial point, executing a DFS algorithm based on a forward graph to search a path, wherein the path 1 is more than 5 and more than 3, the node 3 is a hot point node, and at the moment, the search can be stopped, and the path 1 is more than 5 and more than 3 is taken as a first path. Taking the initial account number node 4 as an initial point, executing a DFS algorithm to search for a path based on a reverse graph (a dotted line part in FIG. 4) of the forward graph, and stopping the search when the path 4 is greater than 2 and greater than 6 and the node 6 is a hot point node, and taking the path 4 greater than 2 and greater than 6 as a second path and taking the reverse path of the second path as 6 greater than 2 and greater than 4. And determining that the hot spot node 3 is used as an initial point and the third path with the hot spot node 6 as a final point is 3 & gt 7 & gt 8 & gt 6 based on the hot spot node path management library. And then, the paths 1 & gt 5 & gt 3, 6 & gt 2 & gt 4, 3 & gt 7 & gt 8 & gt 6 and the paths 4 & gt 1 between the initial account number nodes and the terminal account number nodes are spliced to obtain the annular paths, wherein 1 & gt 5 & gt 3 & gt 7 & gt 8 & gt 6 & gt 2 & gt 4 & gt 1.
Next, the construction of the hotspot node path management library and the determination method of the hotspot node are specifically described.
Determining the heat value of each original node in the forward graph according to the following formula:
Figure 708571DEST_PATH_IMAGE034
wherein the content of the first and second substances,
Figure 496399DEST_PATH_IMAGE035
representing an original nodeuThe value of the heat of the gas (C),
Figure 132916DEST_PATH_IMAGE036
representing an original nodeuThe degree of (c) is determined,
Figure 20101DEST_PATH_IMAGE037
representing an original nodeuThe number of corresponding triangular structures;
and determining the original node with the heat value reaching the heat threshold as the hot point node.
The formula can be known from the formula for determining the heat value of each original node in the forward graph
Figure 46963DEST_PATH_IMAGE038
And
Figure 700798DEST_PATH_IMAGE039
two parts are formed.
Figure 300144DEST_PATH_IMAGE040
Representing an original nodeuThe degree of (c) is the sum of the out degree and the in degree. By passing
Figure 482864DEST_PATH_IMAGE039
The degree of the original node can be measured, the greater the degree of the original node is, the more the paths passing through the original node are, the greater the heat value of the original node is, the logarithm of the degree of the original node is taken because the integral degree distribution of the original node is usually long tail distribution, and the long tail effect of the long tail distribution can be eliminated by adopting the logarithm taking mode, so that the calculation of the heat value is more reasonable and accurate.
Figure 262601DEST_PATH_IMAGE038
The number of node-averaged triangular structures is shown, wherein,
Figure 64335DEST_PATH_IMAGE041
representing the original nodeuCorresponding to the number of triangular structures, and
Figure 409866DEST_PATH_IMAGE042
indicating the combined logarithm that can be generated between the primary nodes connected to the primary node, and, therefore,
Figure 904432DEST_PATH_IMAGE038
the method can be used for representing the density of each node around the original node, and the higher the density is, the higher the possibility that the original node can be traversed is, and the greater the heat value of the original node is.
It should be noted that, when the heat value of the original node is greater than or equal to the heat threshold, the original node is taken as a hot-spot node. The heat threshold may be a threshold parameter set according to actual requirements, and factors affecting the heat threshold include storage capacity of the hotspot node path management library, real-time response time, and the like. If the storage capacity of the hotspot node path management library is larger, the more hotspot node path data can be stored, and the heat threshold value can be reduced, so that more hotspot node path data can be stored conveniently. If the real-time response time requirement is high, more hot point node path data can be stored, the real-time calculation amount is reduced, and the heat threshold value can be adjusted to be small. The specific setting manner of the heat threshold is not particularly limited in this embodiment.
Taking any hot point node as an initial point, and executing a DFS algorithm based on a forward graph to search a passing path between the hot point nodes; and storing the searched traffic path to obtain a hotspot node path management library.
Specifically, any hot point node is taken as an initial point, a DFS algorithm is executed based on a forward graph to search, a passing path between the hot point nodes is determined, and the passing path is stored in a hot point node path management library so as to be used for determining a ring path in the follow-up process.
It should be noted that the advantage of setting the hot spot node path management library is that since the possibility that the path between the hot spot nodes is used in the circular path is high, storing the path between the hot spot nodes in the hot spot node path management library can avoid the waste of time and resources caused by repeatedly performing path search between the hot spot nodes, and can improve the real-time performance of determining the circular path, thereby improving the efficiency of identifying the target group.
In order to ensure the real-time performance of the path data stored in the hotspot node path management library, the path stored in the hotspot node path management library can be updated according to the dynamically updated forward graph.
If the updated forward graph comprises the newly added edge, executing a DFS algorithm to perform path search by taking the terminal node of the newly added edge as an initial point based on the forward graph, stopping the execution of the DFS algorithm when a hot point node is searched, marking the hot point node as a third hot point node, and recording the searched third path; and taking the initial node of the newly added edge as an initial point, executing a DFS algorithm to search a path based on a reverse graph of the forward graph, stopping the execution of the DFS algorithm when a hot point node is searched, marking the hot point node as a fourth hot point node, and recording the searched fourth path. And splicing the newly added edge with the reverse paths of the third path and the fourth path to obtain a path from the fourth hot point node to the third hot point node, and storing the path into a hot point node path management library.
If the updated forward graph includes deleted edges, the paths including the deleted edges in the hotspot node path management library can be deleted. Meanwhile, the degrees of the starting point and the ending point of the deleted edge are reduced by one so as to determine whether the starting point and the ending point of the deleted edge are hot point nodes or not again. And if the starting point and/or the end point of the deleted edge is changed into a non-hotspot node from the hotspot node, deleting the path containing the node in the hotspot node path management library.
S230, determining the aggregation metric value of each account node in the annular path based on the similarity of the set characteristic dimensions among the account nodes in the annular path.
S240, determining whether the user group corresponding to each account node in the annular path is a target group according to the aggregative metric value.
The technical scheme of the embodiment of the invention determines a primary account node initiating a presentation behavior and a presented terminal account node according to the behavior of mutually presenting target props among users, executes a depth-first search DFS algorithm based on a stored forward map according to the primary account node and the terminal account node to obtain a ring-shaped path consisting of the primary account node, at least one intermediate account node and the terminal account node, determines the aggregation metric value of each account node in the ring-shaped path based on the similarity of set characteristic dimensions among the account nodes in the ring-shaped path, further determines whether a user group corresponding to each account node in the ring-shaped path is a target group according to the aggregation metric value, solves the problems of higher complexity and poorer real-time when determining the target group by constructing a complex graph relationship, and realizes accurate identification and search of the target group, and the technical effect of the identification cost is reduced.
The following is an embodiment of the object group identification apparatus provided in the embodiment of the present invention, which belongs to the same inventive concept as the object group identification methods in the above embodiments, and details not described in detail in the embodiment of the object group identification apparatus may refer to the embodiment of the object group identification method.
EXAMPLE III
Fig. 5 is a schematic structural diagram of a target group identification apparatus according to a third embodiment of the present invention, where the apparatus includes: a loop path determination module 310, an aggregative metric value determination module 320, and a target community determination module 330.
The loop path determining module 310 is configured to determine a loop path for transferring a target prop between account nodes according to a behavior of presenting the target prop between users; an aggregation metric value determining module 320, configured to determine an aggregation metric value of each account node in the circular path based on a similarity between the account nodes in the circular path in a set feature dimension; and the target group determining module 330 is configured to determine whether a user group corresponding to each account node in the circular path is a target group according to the aggregative metric value.
Optionally, the circular path determining module 310 is specifically configured to determine, according to a behavior of presenting a target item to each other between users, a primary account node initiating a presenting behavior and a presented terminal account node; according to the initial account number node and the terminal account number node, a depth-first search (DFS) algorithm is executed based on a stored forward graph, and the annular path composed of the initial account number node, at least one intermediate account number node and the terminal account number node is obtained; the forward graph is constructed based on transfer data of the target prop between the account number nodes in a set historical time period.
Optionally, the apparatus further comprises: the forward graph building module is specifically used for building the forward graph according to the following mode based on transfer data of the target prop between account number nodes in a set historical time period: determining that account nodes of target item presentation and presented behaviors exist in a set historical time period, and taking the account nodes as original nodes of the forward graph; and aiming at each original node, if the target prop is transferred from a first original node to a second original node, generating an edge pointing to the second original node from the first original node between the first original node and the second original node, and thus constructing the forward graph.
Optionally, the apparatus further comprises: the forward map updating module is specifically used for dynamically updating the forward map according to set information; the setting information includes at least one of: real-time information transfer of the target prop between account number nodes; logout information of the account number node; and the effective duration of the edges between the original nodes.
Optionally, the annular path determining module 310 is further configured to execute a DFS algorithm based on the forward graph to perform path search by using the terminal account node as an origin, stop the execution of the DFS algorithm when a first set condition is met, and record a searched first path; executing a DFS algorithm to search paths by taking the initial account number node as an initial point based on a reverse graph of the forward graph, stopping the execution of the DFS algorithm when a second set condition is met, and recording the searched second path; and determining the annular path according to the first path and the second path.
Optionally, the first setting condition includes at least one of: the initial account node is traversed to or to a hot point node; the second setting condition includes at least one of: the terminal account node is traversed to or to a hot point node; when the first setting condition and the second setting condition are both traversed to a hotspot node, the annular path determining module 310 is further configured to determine, based on a hotspot node path management library, a third path between a first hotspot node in the first path and a second hotspot node in the second path; and splicing the reverse paths of the first path and the second path and the third path to obtain the annular path.
Optionally, the apparatus further comprises: the hot spot node path management library determining module is used for determining the hot spot node path management library based on the following modes:
determining the heat value of each original node in the forward graph according to the following formula:
Figure 905886DEST_PATH_IMAGE043
wherein the content of the first and second substances,
Figure 635945DEST_PATH_IMAGE035
representing an original nodeuThe value of the heat of the gas (C),
Figure 711348DEST_PATH_IMAGE040
representing an original nodeuThe degree of (c) is determined,
Figure 501450DEST_PATH_IMAGE044
representing an original nodeuThe number of corresponding triangular structures;
determining the original node with the heat value reaching the heat threshold as a hot point node; taking any hot point node as an initial point, and executing a DFS algorithm based on the forward graph to search a passing path between the hot point nodes; and storing the searched traffic path to obtain the hotspot node path management library.
Optionally, the aggregative metric value determining module 320 is specifically configured to determine the aggregative metric value of each account node in the ring path based on the following formula:
Figure 895260DEST_PATH_IMAGE045
wherein the content of the first and second substances,
Figure 632272DEST_PATH_IMAGE046
is an aggregative metric value of each account node in the ring path,
Figure 952394DEST_PATH_IMAGE003
is an account node on the ring pathuAndvin the first placeiThe similarity in the dimension of each set feature,
Figure 523184DEST_PATH_IMAGE047
is the firstiThe similarity weight of the feature dimension is set,
Figure 295968DEST_PATH_IMAGE005
is the firstiA feature vector corresponding to each set feature dimension,cis a set of account nodes in the circular path,Iis a set of characteristics that are characteristic of the image,
Figure 571092DEST_PATH_IMAGE006
is the total number of account nodes in the ring path.
Optionally, the apparatus further comprises: a similarity weight determination module, configured to determine a similarity weight of each set feature dimension based on the following formula:
Figure 621087DEST_PATH_IMAGE048
wherein the content of the first and second substances,
Figure 752991DEST_PATH_IMAGE018
is the firstiThe similarity weight of the feature dimension is set,
Figure 950754DEST_PATH_IMAGE005
is the firstiA feature vector corresponding to each set feature dimension,
Figure 967252DEST_PATH_IMAGE049
is that
Figure 996388DEST_PATH_IMAGE050
The transposed vector of (a) is,
Figure 413375DEST_PATH_IMAGE051
is a first matrix of the degree of similarity,
Figure 364013DEST_PATH_IMAGE052
is a second similarity matrix of the first and second images,
Figure 777677DEST_PATH_IMAGE021
and
Figure 802265DEST_PATH_IMAGE022
is determined according to the following formula
Figure 10392DEST_PATH_IMAGE053
Figure 854852DEST_PATH_IMAGE054
Wherein the content of the first and second substances,
Figure 275469DEST_PATH_IMAGE025
represents the first in the first similarity matrixjGo to the firstkThe value of the column is such that,
Figure 13617DEST_PATH_IMAGE026
representing the second in the second similarity matrixjGo to the firstkThe value of the column is such that,llabels representing the account nodes in the determined target community,nwhich represents the number of account number nodes,
Figure 799171DEST_PATH_IMAGE027
indicating having a labellThe number of account number nodes of (1),
Figure 255560DEST_PATH_IMAGE028
is shown asjThe label of each account number node is used,
Figure 384928DEST_PATH_IMAGE029
is shown askLabels of individual account nodes.
According to the technical scheme of the embodiment of the invention, the annular path of the target prop transferred between the account number nodes is determined according to the behavior of the target prop given by the users mutually, the aggregation metric value of each account number node in the annular path is determined based on the similarity of the set characteristic dimension between the account number nodes in the annular path, and then whether the user group corresponding to each account number node in the annular path is the target group is determined according to the aggregation metric value, so that the problems of high complexity and poor real-time performance when the target group is determined by constructing a complex graph relation are solved, the target group is accurately identified and searched, and the technical effect of identification cost is reduced.
The target group identification device provided by the embodiment of the invention can execute the target group identification method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.
It should be noted that, the units and modules included in the apparatus are merely divided according to functional logic, but are not limited to the above division as long as the corresponding functions can be implemented; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the embodiment of the invention.
Example four
Fig. 6 is a schematic structural diagram of an electronic device according to a fourth embodiment of the present invention. FIG. 6 illustrates a block diagram of an exemplary electronic device 40 suitable for use in implementing embodiments of the present invention. The electronic device 40 shown in fig. 6 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiment of the present invention.
As shown in fig. 6, electronic device 40 is embodied in the form of a general purpose computing device. The components of electronic device 40 may include, but are not limited to: one or more processors or processing units 401, a system memory 402, and a bus 403 that couples the various system components (including the system memory 402 and the processing unit 401).
Bus 403 represents one or more of any of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus using any of a variety of bus architectures. By way of example, such architectures include, but are not limited to, Industry Standard Architecture (ISA) bus, micro-channel architecture (MAC) bus, enhanced ISA bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.
Electronic device 40 typically includes a variety of computer system readable media. Such media may be any available media that is accessible by electronic device 40 and includes both volatile and nonvolatile media, removable and non-removable media.
The system memory 402 may include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM) 404 and/or cache memory 405. The electronic device 40 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 406 may be used to read from and write to non-removable, nonvolatile magnetic media (not shown in FIG. 6, commonly referred to as a "hard drive"). Although not shown in FIG. 6, a magnetic disk drive for reading from and writing to a removable, nonvolatile magnetic disk (e.g., a "floppy disk") and an optical disk drive for reading from or writing to a removable, nonvolatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 403 by one or more data media interfaces. System memory 402 may include at least one program product having a set (e.g., at least one) of program modules that are configured to carry out the functions of embodiments of the invention.
A program/utility 408 having a set (at least one) of program modules 407 may be stored, for example, in system memory 402, such program modules 407 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which examples or some combination thereof may comprise an implementation of a network environment. Program modules 407 generally perform the functions and/or methods of the described embodiments of the invention.
The electronic device 40 may also communicate with one or more external devices 409 (e.g., keyboard, pointing device, display 410, etc.), with one or more devices that enable a user to interact with the electronic device 40, and/or with any devices (e.g., network card, modem, etc.) that enable the electronic device 40 to communicate with one or more other computing devices. Such communication may be through input/output (I/O) interface 411. Also, the electronic device 40 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network, such as the internet) via the network adapter 412. As shown, the network adapter 412 communicates with the other modules of the electronic device 40 over the bus 403. It should be appreciated that although not shown in FIG. 6, other hardware and/or software modules may be used in conjunction with electronic device 40, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.
The processing unit 401 executes various functional applications and data processing by running the program stored in the system memory 402, for example, to implement the steps of the target community identification method provided by the embodiment of the present invention, the method including:
determining a circular path for transferring the target prop between account number nodes according to the behavior of presenting the target prop mutually among users;
determining an aggregation metric value of each account node in the annular path based on the similarity of the set characteristic dimensions among the account nodes in the annular path;
and determining whether a user group corresponding to each account node in the annular path is a target group according to the aggregative metric value.
Of course, those skilled in the art can understand that the processor can also implement the technical solution of the target user identification method provided by any embodiment of the present invention.
EXAMPLE five
Fifth embodiment of the present invention further provides a storage medium containing computer-executable instructions, which when executed by a computer processor, are configured to perform the steps of the target community identification method provided in any of the embodiments of the present invention, the method including:
determining a circular path for transferring the target prop between account number nodes according to the behavior of presenting the target prop mutually among users;
determining an aggregation metric value of each account node in the annular path based on the similarity of the set characteristic dimensions among the account nodes in the annular path;
and determining whether a user group corresponding to each account node in the annular path is a target group according to the aggregative metric value.
Computer storage media for embodiments of the invention may employ any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. The computer-readable storage medium may be, for example but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
It will be understood by those skilled in the art that the modules or steps of the invention described above may be implemented by a general purpose computing device, they may be centralized on a single computing device or distributed across a network of computing devices, and optionally they may be implemented by program code executable by a computing device, such that it may be stored in a memory device and executed by a computing device, or it may be separately fabricated into various integrated circuit modules, or it may be fabricated by fabricating a plurality of modules or steps thereof into a single integrated circuit module. Thus, the present invention is not limited to any specific combination of hardware and software.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims (9)

1. A target community identification method, comprising:
determining a circular path for transferring the target prop between account number nodes according to the behavior of presenting the target prop mutually among users;
determining an aggregation metric value of each account node in the annular path based on the similarity of the set characteristic dimensions among the account nodes in the annular path;
determining whether a user group corresponding to each account node in the annular path is a target group or not according to the aggregative metric value;
the determining of the annular path of the target prop transferred between the account number nodes according to the behavior of mutual presentation of the target prop between users comprises the following steps:
determining an initial account node initiating a presentation behavior and a presented terminal account node according to the behavior of presenting target props among users;
according to the initial account number node and the terminal account number node, a depth-first search (DFS) algorithm is executed based on a stored forward graph, and the annular path composed of the initial account number node, at least one intermediate account number node and the terminal account number node is obtained;
the forward graph is constructed based on transfer data of the target prop between the account number nodes in a set historical time period.
2. The method according to claim 1, wherein the forward graph is constructed based on transfer data of target props among account number nodes in a set historical time period as follows:
determining that account nodes of target item presentation and presented behaviors exist in a set historical time period, and taking the account nodes as original nodes of the forward graph;
and aiming at each original node, if the target prop is transferred from a first original node to a second original node, generating an edge pointing to the second original node from the first original node between the first original node and the second original node, and thus constructing the forward graph.
3. The method of claim 2, further comprising:
dynamically updating the forward graph according to set information;
the setting information includes at least one of:
real-time information transfer of the target prop between account number nodes;
logout information of the account number node;
and the effective duration of the edges between the original nodes.
4. The method according to claim 1, wherein the performing a depth-first search (DFS) algorithm based on the stored forward graph according to the primary account node and the terminal account node to obtain the circular path composed of the primary account node, at least one intermediate account node, and the terminal account node comprises:
executing a DFS algorithm to search paths by taking the terminal account number node as an initial point based on the digraph, stopping the execution of the DFS algorithm when a first set condition is met, and recording the searched first path;
executing a DFS algorithm to search paths by taking the initial account number node as an initial point based on a reverse graph of the forward graph, stopping the execution of the DFS algorithm when a second set condition is met, and recording the searched second path;
and determining the annular path according to the first path and the second path.
5. The method according to claim 4, wherein the first set condition includes at least one of: the initial account node is traversed to or to a hot point node;
the second setting condition includes at least one of: the terminal account node is traversed to or to a hot point node;
when the first setting condition and the second setting condition are both traversed to a hotspot node, the determining the annular path according to the first path and the second path includes:
determining a third path between a first hotspot node in the first path and a second hotspot node in the second path based on a hotspot node path management library;
and splicing the reverse paths of the first path and the second path and the third path to obtain the annular path.
6. The method of claim 5, wherein the hotspot node path management library is determined based on:
determining the heat value of each original node in the forward graph according to the following formula:
Figure DEST_PATH_IMAGE001
wherein the content of the first and second substances,
Figure 318447DEST_PATH_IMAGE002
representing an original nodeuThe value of the heat of the gas (C),
Figure DEST_PATH_IMAGE003
representing an original nodeuThe degree of (c) is determined,
Figure 438850DEST_PATH_IMAGE004
representing an original nodeuThe number of corresponding triangular structures;
determining the original node with the heat value reaching the heat threshold as a hot point node;
taking any hot point node as an initial point, and executing a DFS algorithm based on the forward graph to search a passing path between the hot point nodes;
and storing the searched traffic path to obtain the hotspot node path management library.
7. The method according to any one of claims 1 to 6, wherein the determining the aggregation metric value of each account node in the circular path based on the similarity between each account node in the circular path in a set feature dimension comprises:
determining an aggregability metric value of each account node in the annular path based on the following formula:
Figure DEST_PATH_IMAGE005
wherein the content of the first and second substances,
Figure 546483DEST_PATH_IMAGE006
is an aggregative metric value of each account node in the ring path,
Figure DEST_PATH_IMAGE007
is an account node on the ring pathuAndvin the first placeiThe similarity in the dimension of each set feature,
Figure 521786DEST_PATH_IMAGE008
is the firstiThe similarity weight of the feature dimension is set,
Figure DEST_PATH_IMAGE009
is the firstiA feature vector corresponding to each set feature dimension,cis a set of account nodes in the circular path,Iis a set of characteristics that are characteristic of the image,
Figure 664054DEST_PATH_IMAGE010
is the total number of account nodes in the ring path.
8. The method of claim 7, wherein the similarity weight for each set feature dimension is determined based on the following formula:
Figure DEST_PATH_IMAGE011
wherein the content of the first and second substances,
Figure 689779DEST_PATH_IMAGE012
is the firstiThe similarity weight of the feature dimension is set,
Figure DEST_PATH_IMAGE013
is the firstiA feature vector corresponding to each set feature dimension,
Figure 35441DEST_PATH_IMAGE014
is that
Figure DEST_PATH_IMAGE015
The transposed vector of (a) is,
Figure 139401DEST_PATH_IMAGE016
is a first matrix of the degree of similarity,
Figure DEST_PATH_IMAGE017
is a second similarity matrix of the first and second images,
Figure 11542DEST_PATH_IMAGE018
and
Figure DEST_PATH_IMAGE019
is determined according to the following formula
Figure 332802DEST_PATH_IMAGE020
Figure DEST_PATH_IMAGE021
Wherein the content of the first and second substances,
Figure 165760DEST_PATH_IMAGE022
represents the first in the first similarity matrixjGo to the firstkThe value of the column is such that,
Figure 637192DEST_PATH_IMAGE023
representing the second in the second similarity matrixjGo to the firstkThe value of the column is such that,llabels representing the account nodes in the determined target community,nwhich represents the number of account number nodes,
Figure 754053DEST_PATH_IMAGE024
indicating having a labellThe number of account number nodes of (1),
Figure DEST_PATH_IMAGE025
is shown asjThe label of each account number node is used,
Figure 652739DEST_PATH_IMAGE026
is shown askLabels of individual account nodes.
9. An electronic device, characterized in that the electronic device comprises:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement the target community identification method of any one of claims 1-8.
CN202110248766.1A 2021-03-08 2021-03-08 Target group identification method and electronic equipment Active CN112616074B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110248766.1A CN112616074B (en) 2021-03-08 2021-03-08 Target group identification method and electronic equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110248766.1A CN112616074B (en) 2021-03-08 2021-03-08 Target group identification method and electronic equipment

Publications (2)

Publication Number Publication Date
CN112616074A CN112616074A (en) 2021-04-06
CN112616074B true CN112616074B (en) 2021-05-28

Family

ID=75254393

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110248766.1A Active CN112616074B (en) 2021-03-08 2021-03-08 Target group identification method and electronic equipment

Country Status (1)

Country Link
CN (1) CN112616074B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113553498A (en) * 2021-06-18 2021-10-26 北京旷视科技有限公司 Community updating method, apparatus, electronic device and computer readable medium

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108875817A (en) * 2018-06-06 2018-11-23 网易（杭州）网络有限公司 Identify plug-in method and device, storage medium, electronic device
CN109464807A (en) * 2018-11-06 2019-03-15 网易（杭州）网络有限公司 Detect game plug-in method, apparatus and terminal
CN110177094A (en) * 2019-05-22 2019-08-27 武汉斗鱼网络科技有限公司 A kind of user community recognition methods, device, electronic equipment and storage medium
US10463953B1 (en) * 2013-07-22 2019-11-05 Niantic, Inc. Detecting and preventing cheating in a location-based game
CN111311416A (en) * 2020-02-28 2020-06-19 杭州云象网络技术有限公司 Block chain money laundering node detection method based on multichannel graph and graph neural network
US10881964B1 (en) * 2018-09-13 2021-01-05 Electronic Arts Inc. Automated detection of emergent behaviors in interactive agents of an interactive environment
CN112206500A (en) * 2019-05-20 2021-01-12 天使游戏纸牌股份有限公司 Management system for game coins
CN112370793A (en) * 2020-11-25 2021-02-19 上海幻电信息科技有限公司 Risk control method and device for user account

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10357713B1 (en) * 2016-08-05 2019-07-23 Wells Fargo Bank, N.A. Utilizing gaming behavior to evaluate player traits
CN107517394A (en) * 2017-09-01 2017-12-26 北京小米移动软件有限公司 Identify the method, apparatus and computer-readable recording medium of disabled user
CN108776892B (en) * 2018-05-21 2022-05-31 北京橙鑫数据科技有限公司 Storage system, device, and recovery method of storage system
CN109464808A (en) * 2018-11-06 2019-03-15 网易（杭州）网络有限公司 Detect game plug-in method, apparatus and terminal

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10463953B1 (en) * 2013-07-22 2019-11-05 Niantic, Inc. Detecting and preventing cheating in a location-based game
CN108875817A (en) * 2018-06-06 2018-11-23 网易（杭州）网络有限公司 Identify plug-in method and device, storage medium, electronic device
US10881964B1 (en) * 2018-09-13 2021-01-05 Electronic Arts Inc. Automated detection of emergent behaviors in interactive agents of an interactive environment
CN109464807A (en) * 2018-11-06 2019-03-15 网易（杭州）网络有限公司 Detect game plug-in method, apparatus and terminal
CN112206500A (en) * 2019-05-20 2021-01-12 天使游戏纸牌股份有限公司 Management system for game coins
CN110177094A (en) * 2019-05-22 2019-08-27 武汉斗鱼网络科技有限公司 A kind of user community recognition methods, device, electronic equipment and storage medium
CN111311416A (en) * 2020-02-28 2020-06-19 杭州云象网络技术有限公司 Block chain money laundering node detection method based on multichannel graph and graph neural network
CN112370793A (en) * 2020-11-25 2021-02-19 上海幻电信息科技有限公司 Risk control method and device for user account

Also Published As

Publication number Publication date
CN112616074A (en) 2021-04-06

Similar Documents

Publication Publication Date Title
CN110278175B (en) Graph structure model training and garbage account identification method, device and equipment
CN106940705B (en) Method and equipment for constructing user portrait
CN110750654A (en) Knowledge graph acquisition method, device, equipment and medium
CN110909540B (en) Method and device for identifying new words of short message spam and electronic equipment
WO2019169766A1 (en) Electronic apparatus, method and system for early warning regarding system sensitive content, and storage medium
CN103870528A (en) Method and system for question classification and feature mapping in deep question answering system
CN112616074B (en) Target group identification method and electronic equipment
CN115034596A (en) Risk conduction prediction method, device, equipment and medium
CN111291936B (en) Product life cycle prediction model generation method and device and electronic equipment
CN114090155A (en) Robot process automation interface element positioning method and device and storage medium
CN111475736A (en) Community mining method, device and server
CN108319688A (en) A kind of method and apparatus for user read prompting
CN111563527B (en) Abnormal event detection method and device
CN110598122A (en) Social group mining method, device, equipment and storage medium
CN110363583B (en) Method and device for creating user consumption behavior label based on position information and electronic equipment
CN113365113B (en) Target node identification method and device
CN113723890B (en) Information processing method, device, equipment and storage medium
CN116127154A (en) Knowledge tag recommendation method and device, electronic equipment and storage medium
CN111125272B (en) Regional characteristic acquisition method, regional characteristic acquisition device, computer equipment and medium
US11849342B2 (en) Global and local measures of centrality for signed and unsigned networks
CN114925050A (en) Data verification method and device based on knowledge base, electronic equipment and storage medium
CN114357180A (en) Knowledge graph updating method and electronic equipment
CN111598714A (en) Two-stage unsupervised group partner identification method and device and electronic equipment
CN110750569A (en) Data extraction method, device, equipment and storage medium
CN112261484B (en) Target user identification method and device, electronic equipment and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant