CN109598563A - Brush single detection method, device, storage medium and electronic equipment - Google Patents
Brush single detection method, device, storage medium and electronic equipment Download PDFInfo
- Publication number
- CN109598563A CN109598563A CN201910069412.3A CN201910069412A CN109598563A CN 109598563 A CN109598563 A CN 109598563A CN 201910069412 A CN201910069412 A CN 201910069412A CN 109598563 A CN109598563 A CN 109598563A
- Authority
- CN
- China
- Prior art keywords
- node
- network
- community
- weight
- user
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0207—Discounts or incentives, e.g. coupons or rebates
- G06Q30/0225—Avoiding frauds
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
This disclosure relates to a kind of single detection method of brush, device, storage medium and electronic equipment, to solve to carry out brushing single detection to the independent brush single act of known mode in the related technology, the technical problem that the applicability of detection is not high and detection range is small, this method comprises: establishing related network according to the interbehavior data between user and businessman, the related network includes the multiple summits of multiple nodes and the above-mentioned multiple nodes of connection, which is incidentally used to describe the weight of the tightness degree of the interbehavior between user and businessman；The related network is divided by multiple Web Communities according to the weight of above-mentioned multiple summits by preset label propagation algorithm；Determine that there are the target network communities of brush single act according to the modularity of the Web Community and preset modularity threshold value.The a variety of brush single acts occurred in the form of clique, community can be identified and be detected, the applicability that brush singly detects is improved, expand detection range by the related network that user and businessman form to brush single act.
Description
Technical field
This disclosure relates to payment technical field on line, and in particular, to a kind of single detection method of brush, device, storage medium and electricity
Sub- equipment.
Background technique
With becoming increasingly popular for platform is consumed on line, the illegal brush single act consumed on platform on line is also more and more rampant.
Brush single act include businessman staff or businessman through committing others carry out to consumed on line the commodity on platform or service into
The behavior not for the purpose of normally consuming such as consumption is made a big purchase and verified to row in large quantities.By brush single act, the available falseness of businessman
Sales volume and not occur real consumption in the case where commodity or service are evaluated, conceal or forge key transaction
Information, and then the illegal profits such as the preferential subsidy that consumption platform provides in line taking are covered by false Transaction Information.This behavior
Violate the trust delegation relationship consumed between platform on trade company and line, also in serious infringement platform consumer equity.Therefore, have
Effect, accurate list testing mechanism of brushing improve marketing efficiency with vital for consuming platform maintenance transaction order on line
Value.
Summary of the invention
It is a general object of the present disclosure to provide a kind of single detection method of brush, device, storage medium and electronic equipments, to solve
In the related technology the independent brush single act of known mode can only be carried out brushing single detection, the applicability of detection is not high and detects model
Enclose small technical problem.
To achieve the goals above, disclosure first aspect provides a kind of single detection method of brush, which comprises
Related network is established according to the interbehavior data between user and businessman, the related network includes multiple nodes
And the multiple summits of the multiple node of connection, the side is incidentally used to describe the weight of the tightness degree of the interbehavior；
The related network is divided by multiple nets according to the weight of the multiple summits by preset label propagation algorithm
Network community；
Determine that there are the target networks of brush single act according to the modularity of the Web Community and preset modularity threshold value
Community.
Optionally, the interbehavior data according between user and businessman establish related network, comprising:
Using the User ID of the user as user node, using the Merchant ID of the businessman as businessman's node, to obtain
The multiple node；
According to the interbehavior data, there are establish subsidiary power between the user node of interbehavior and businessman's node
The side of weight, to establish the related network.
Optionally, the interbehavior data include the interactive operation and every kind of interaction behaviour that the interbehavior includes
Make corresponding weighting coefficient and frequency of occurrence, it is described according to the interbehavior data, there are the user nodes of interbehavior
The side of subsidiary weight is established, between businessman's node to establish the related network, comprising:
There are side is established between the user node of interbehavior and businessman's node, with the determination multiple summits；
The one or more target interactive operations for including in the corresponding interbehavior in the first side are obtained, first side is institute
State any bar side in multiple summits；
The summation of the corresponding operation weight of one or more target interactive operations is calculated, the operation weight is described
The product of the frequency of occurrence of target interactive operation weighting coefficient corresponding with the target interactive operation；
It is subsidiary with each edge in the determination multiple summits using the summation of the operation weight as the weight on first side
Weight.
Optionally, described to be drawn the related network according to the weight of the multiple summits by preset label propagation algorithm
It is divided into multiple Web Communities, comprising:
Different labels is added for each node in the related network；
According to the weight on the side between each node and multiple neighbor nodes of each node, pass through the mark
Label relay algorithm and are updated operation to the label of each node；
After completing the operation of the update to the label of each node, it is same for will be provided with the node division of same label
Web Community, to obtain the multiple Web Community.
Optionally, the power according to the side between each node and multiple neighbor nodes of each node
Weight, is updated operation to the label of each node, comprising:
Determine that multiple target base nodes of destination node, the destination node are any section in the related network
Point；
Target base node division by same label subsidiary in the multiple target base node is a node group, with
Obtain the corresponding multiple node groups of the destination node；
Determine have the group of destination nodes of maximum label weight in the multiple node group, the label weight is described
The summation of the weight on the side between all nodes in destination node and each node group；
It is the second label by the first tag update of the destination node, second label is in the group of destination nodes
The label of any node.
Optionally, described to determine that there are brush single acts according to the modularity of the Web Community and preset modularity threshold value
Target network community, comprising:
The adjacent node of each node in first network community is obtained, the adjacent node is the neighbour of each node
The neighbor node that the first network community is not belonging in node is occupied, the first network community is in the multiple Web Community
Any Web Community；
Corresponding second Web Community, the first network community is obtained, second Web Community includes first net
The adjacent node of all nodes and all nodes in network community；
Obtain second while second Web Community include it is all while in shared ratio, as the first network
The object module degree of community, wherein the node at second side both ends belongs to the first network community；
If the object module degree is greater than the modularity threshold value, determine that the first network community is the target network
Community.
Disclosure second aspect provides a kind of single detection device of brush, and described device includes:
Network establishes module, for establishing related network, the pass according to the interbehavior data between user and businessman
Network of networking includes the multiple summits of multiple nodes and the multiple node of connection, the side be incidentally used to describe user and businessman it
Between interbehavior tightness degree weight；
Community's division module, for by preset label propagation algorithm according to the weights of the multiple summits by the association
Network is divided into multiple Web Communities；
Single determining module is brushed, for determining there is brush according to the modularity of the Web Community and preset modularity threshold value
The target network community of single act.
Optionally, the network establishes module, comprising:
Node acquisition submodule, for using the User ID of the user as user node, by the Merchant ID of the businessman
As businessman's node, to obtain the multiple node；
Network setting up submodule, for according to the interbehavior data, there are the user node of interbehavior and quotient
The side of subsidiary weight is established between family's node, to establish the related network.
Optionally, the interbehavior data include the interactive operation and every kind of interaction behaviour that the interbehavior includes
Make corresponding weighting coefficient and frequency of occurrence, the network setting up submodule is used for:
There are side is established between the user node of interbehavior and businessman's node, with the determination multiple summits；
The one or more target interactive operations for including in the corresponding interbehavior in the first side are obtained, first side is institute
State any bar side in multiple summits；
The summation of the corresponding operation weight of one or more target interactive operations is calculated, the operation weight is described
The product of the frequency of occurrence of target interactive operation weighting coefficient corresponding with the target interactive operation；
It is subsidiary with each edge in the determination multiple summits using the summation of the operation weight as the weight on first side
Weight.
Optionally, community's division module, comprising:
Label adds submodule, for adding different labels for each node in the related network；
Tag update submodule, for according between each node and multiple neighbor nodes of each node
The weight on side relays algorithm by the label and is updated operation to the label of each node；
Community divides submodule, for will be provided with identical after completing the operation of the update to the label of each node
The node division of label is consolidated network community, to obtain the multiple Web Community.
Optionally, the tag update submodule, is used for:
Determine that multiple target base nodes of destination node, the destination node are any section in the related network
Point；
Target base node division by same label subsidiary in the multiple target base node is a node group, with
Obtain the corresponding multiple node groups of the destination node；
Determine have the group of destination nodes of maximum label weight in the multiple node group, the label weight is described
The summation of the weight on the side between all nodes in destination node and each node group；
It is the second label by the first tag update of the destination node, second label is in the group of destination nodes
The label of any node.
Optionally, the single determining module of the brush, comprising:
Node determines submodule, for determining the adjacent node of each node in first network community, the adjacent section
Put the neighbor node that the first network community is not belonging in the neighbor node for each node, the first network community
For any Web Community in the multiple Web Community；
Community's acquisition submodule, for obtaining corresponding second Web Community, the first network community, second net
Network community includes the adjacent node of all nodes and all nodes in the first network community；
Modularity acquisition submodule, for obtain second while second Web Community include it is all while in it is shared
Ratio, the object module degree as the first network community, wherein the node at second side both ends belongs to described first
Web Community；
Brush is single to determine submodule, if being greater than the modularity threshold value for the object module degree, determines first net
Network community is the target network community.
The disclosure third aspect provides a kind of computer readable storage medium, is stored thereon with computer program, the program
The step of brush list detection method described in first aspect is realized when being executed by processor.
Disclosure fourth aspect provides a kind of electronic equipment, comprising:
Memory is stored thereon with computer program；
Processor, for executing the computer program in the memory, to realize brush list described in first aspect
The step of detection method.
Using technical solution provided by the present disclosure, following technical effect at least can achieve:
Establish related network according to the interbehavior data between user and businessman, the related network include multiple nodes with
And the multiple summits of the above-mentioned multiple nodes of connection, which is incidentally used to describe the tightness degree of the interbehavior between user and businessman
Weight；The related network is divided by multiple network societies according to the weight of above-mentioned multiple summits by preset label propagation algorithm
Area；Determine that there are the target network communities of brush single act according to the modularity of the Web Community and preset modularity threshold value.Energy
The related network being enough made up of user and businessman to brush single act, to a variety of brush single acts occurred in the form of clique, community into
Row identification and detection improve the applicability that brush singly detects, and expand detection range.
Other feature and advantage of the disclosure will the following detailed description will be given in the detailed implementation section.
Detailed description of the invention
Attached drawing is and to constitute part of specification for providing further understanding of the disclosure, with following tool
Body embodiment is used to explain the disclosure together, but does not constitute the limitation to the disclosure.In the accompanying drawings:
Fig. 1 is a kind of flow chart for brushing single detection method shown according to an exemplary embodiment；
Fig. 2 is according to a kind of flow chart of the method for building up of related network shown in fig. 1；
Fig. 3 is according to a kind of flow chart of the division methods of Web Community shown in fig. 1；
Fig. 4 is according to a kind of single flow chart for determining method of brush shown in fig. 1；
Fig. 5 is a kind of schematic diagram of related network shown according to an exemplary embodiment；
Fig. 6 is a kind of block diagram for brushing single detection device shown according to an exemplary embodiment；
Fig. 7 is the block diagram that module is established according to a kind of network shown in Fig. 6；
Fig. 8 is the block diagram according to a kind of community's division module shown in Fig. 6；
Fig. 9 is the block diagram according to a kind of brush list determining module shown in Fig. 6；
Figure 10 is the structural schematic diagram of a kind of electronic equipment shown according to an exemplary embodiment.
Specific embodiment
It is described in detail below in conjunction with specific embodiment of the attached drawing to the disclosure.It should be understood that this place is retouched
The specific embodiment stated is only used for describing and explaining the disclosure, is not limited to the disclosure.
Platform is consumed on line usually by based on the detection such as simple statistics, regulation engine or Supervised machine learning model
Mode detects brush single act.Wherein, the detection mode based on simple statistics and regulation engine is only related to businessman
The quantity of trading activity or the changing rule of quantity are counted, relatively low to the discrimination of brush single act.Meanwhile brushing list person
It is easy to evade this detection mode by the characteristic of simple statistics or regulation engine itself.
In addition, needing in the detection mode based on Supervised machine learning model for known one or more of brushes
Single act acquires a large amount of sample data, then is trained by these sample datas to machine learning model, and then pass through instruction
The model perfected identifies this one or more of brush single act.But illegal activities are usually all advanced, brush single acts
It is also in this way, brush single mode is also being constantly updated, and gradually forms the brush single mode of clique's type with scientific and technical continuous upgrading
Formula, and above-mentioned Supervised machine learning model can only be examined for the independent brush single act for having known brush single mode
It surveys, the applicability of detection is not high, and detection range is also smaller.
Inventor notices this problem, proposes a kind of new brush list detection method, specific as follows:
Fig. 1 is a kind of flow chart for brushing single detection method shown according to an exemplary embodiment, as shown in Figure 1, the party
Method the following steps are included:
In a step 101, related network is established according to the interbehavior data between user and businessman.
Wherein, which includes the multiple summits of multiple nodes and the above-mentioned multiple nodes of connection, which is user
Node or businessman's node, the side are incidentally used to describe the weight of the tightness degree of the interbehavior between user and businessman.It can be with
Understand, which is a user-businessman's bigraph (bipartite graph) structure, that is, the node in the related network is divided into two classes (i.e.
User node and businessman's node), and the nonoriented edge of subsidiary weight can only be present between user node and businessman's node.Namely
Say, be not in the related network connect two user nodes while or two businessman's nodes of connection while.The interaction row
Being includes a variety of interactive operations between user and businessman, for example, the behaviour that user browses businessman, places an order or comments on
Reply, delivery or the operation for confirming reimbursement that work or businessman carry out user.
In a step 102, the related network is drawn according to the weight of above-mentioned multiple summits by preset label propagation algorithm
It is divided into multiple Web Communities.
Illustratively, existing label propagation algorithm includes two stages: being every in the related network in initial phase
A node adds a unique label；Hereafter, then in the tag update stage update the label of each node, that is, enable a section
Point label is identical as the label of most of neighbours of the node, until the label of each node no longer changes.In disclosure reality
Applying is the concept that joined weight in the above-mentioned tag update stage in the step 102 of example, that is, when updating label, not only to be examined
The quantity for having the neighbor node of same label is considered, it is also contemplated that the weight on the side between node and each neighbor node.In addition,
It should be noted that above-mentioned multiple Web Communities are mutual exclusions, there is no the relationships for including mutually.
In step 103, according in the Web Community modularity and preset modularity threshold value determine there are brush unilines
For target network community.
Illustratively, after obtaining the target network community, all user nodes pair in the target network community can be determined
The user answered is brush single user.
In conclusion technical solution provided by embodiment of the disclosure, it can be according to the interaction between user and businessman
Behavioral data establishes related network, which includes the multiple summits of multiple nodes and the above-mentioned multiple nodes of connection, the side
It is attached to the weight for describing the tightness degree of the interbehavior between user and businessman；Pass through preset label propagation algorithm root
The related network is divided into multiple Web Communities according to the weight of above-mentioned multiple summits；According to the modularity of the Web Community and preset
Modularity threshold value determine there are the target network communities of brush single act.The related network pair that can be made up of user and businessman
Brush single act is identified and is detected to a variety of brush single acts occurred in the form of clique, community, and what raising brush singly detected is applicable in
Property, expand detection range.
Fig. 2 is according to a kind of flow chart of the method for building up of related network shown in fig. 1, as shown in Fig. 2, the step 101
May include:
In step 1011, using the User ID of user as user node, using the Merchant ID of businessman as businessman's node, with
Obtain above-mentioned multiple nodes.
Illustratively, it before above-mentioned steps 1011, can be extracted from the background system log for consuming platform on line all
The User ID of user, the Merchant ID of all businessmans and above-mentioned interbehavior data.
In step 1012, according to the interbehavior data, there are the user node of interbehavior and businessman's node it
Between establish the side of subsidiary weight, to establish the related network.
Illustratively, which includes the interactive operation and every kind of interactive operation pair that the interbehavior includes
The weighting coefficient and frequency of occurrence answered, the step 1012 include: there are between the user node of interbehavior and businessman's node
Side is established, to determine the multiple summits；The one or more target interactive operations for including in the corresponding interbehavior in the first side are obtained,
This is first while for any bar in the multiple summits；Calculate the total of the corresponding operation weight of the one or more target interactive operations
With the product for the frequency of occurrence weighting coefficient corresponding with the target interactive operation that, the operation weight is the target interactive operation；
Using the summation of the operation weight as the weight on first side, with the subsidiary weight of each edge in the determining multiple summits.Specifically,
For example, user A includes: click to the interbehavior between businessman B, browses, places an order and comment on these four interactive operations, correspond to
Weighting coefficient be respectively a1, a2, a3 and a4, also, click and have occurred 5 times, browsing has occurred 3 times, place an order and have occurred 2 times,
Comment has occurred 1 time.Then the corresponding operation weight of these four interactive operations is respectively 5*a1,3*a2,2*a3 and 1*a4.It counts again
Calculate 5*a1, the summation of 3*a2,2*a3 and 1*a4, as user node A to the weight on the side between businessman's node B.
Fig. 3 is according to a kind of flow chart of the division methods of Web Community shown in fig. 1, as shown in figure 3, the step 102
May include:
In step 1021, different labels is added for each node in the related network.
In step 1022, according to the power on the side between above-mentioned each node and multiple neighbor nodes of above-mentioned each node
Weight relays algorithm by the label and is updated operation to the label of above-mentioned each node.
Illustratively, which comprises determining that multiple target base nodes of destination node, the destination node are the pass
Any node in networking network；Target base node division by same label subsidiary in above-mentioned multiple target base nodes is one
A node group, to obtain the corresponding multiple node groups of the destination node；Determine in above-mentioned multiple node groups have maximum label
The group of destination nodes of weight, the label weight are the weight on the side between all nodes in the destination node and each node group
Summation；It is the second label by the first tag update of the destination node, which is any section in the group of destination nodes
The label of point.Specifically, for example, node A is attached to label 1, multiple target base nodes of node A include: node B, node C and
Node D.Wherein, the label of node B is label 2, is divided into node group 1, node C and node D label are all label 3, by it
Node group 2 is divided into, also, the weight on side of the weight on the side between node A and node B between 10, with node C is 5, with section
The weight on the side between point D is 6.In this way, the corresponding label weight of node group 1 is 10, the corresponding label weight of node group 2 is 5+
6=11.It is it is possible to further determine that node group 2 is to have the group of destination nodes of maximum label weight, node A is original
Label 1 replaces with label 3.In step 1022, can through the above steps the label to each node in related network into
Row updates, until the label of each node in related network reaches stable state.
In step 1023, after completing the operation of the update to the label of above-mentioned each node, same label will be provided with
Node division is consolidated network community, to obtain above-mentioned multiple Web Communities.
Fig. 4 is according to a kind of single flow chart for determining method of brush shown in fig. 1, as shown in figure 4, the step 103 can wrap
It includes:
In step 1031, the adjacent node of each node in first network community is obtained.
Wherein, which saves to be not belonging to the neighbours of the first network community in the neighbor node of above-mentioned each node
Point, the first network community are any Web Community in above-mentioned multiple Web Communities.
In step 1032, corresponding second Web Community, the first network community is obtained.
Wherein, which includes the neighbour of all nodes and above-mentioned all nodes in the first network community
Connect node.
In step 1033, obtain second while second Web Community include it is all while in shared ratio, as
The object module degree of the first network community.
Wherein, the node at the second side both ends belongs to the first network community.
Illustratively, in actual program development process, the concept of adjacency matrix can be introduced into the section in Web Community
Point and side carry out digitization statistics.Specifically, it is corresponding with second Web Community that the first network community can be obtained respectively
Adjacency matrix, and above-mentioned object module degree is calculated by modularity calculation formula.The modularity calculation formula can be expressed as
Following equation (1):
Wherein, M is the object module degree, Ai,jFor corresponding first adjacency matrix in the first network community, i and j) representative
Any two node of first adjacency matrix, Lu,vCorresponding second adjacency matrix in the first network community, u and v indicate this
Any two node of two adjacency matrix.Second adjacency matrix can be set are as follows:
By above-mentioned setting rule, can get only include 1 and 0 the second Adjacency matrix, wherein the quantity of " 1 " is
For the quantity on all sides that second Web Community includes.
In step 1034, if the object module degree is greater than the modularity threshold value, determine that the first network community is the mesh
Mark Web Community.
Fig. 5 is a kind of schematic diagram of related network shown according to an exemplary embodiment, as shown in figure 5, dotted line in figure
The part surrounded is the Web Community X that 1021-1023 is marked off through the above steps, it can be seen that Web Community X packet
Containing 7 nodes, wherein the node in title comprising " A " is user node, and the node comprising " B " is businessman's node in title.From
As can be seen that the adjacent node of Web Community X includes node A5 and node B3 in figure.It is then available to arrive Web Community Y, the net
Network community Y includes: all nodes of node A5, node B3 and Web Community X.After this, the second side in Web Community Y
The quantity on (i.e. the side that the node at both ends belongs to Web Community X) is 10, and the total quantity on the side Web Community Y is 12, then may be used
To determine the object module degree of Web Community X for 5/6.When above-mentioned modularity threshold value is 0.5, Web Community X is determined
To there is the high target network community for brushing single risk, and then determine user representated by each user node in Web Community X
It is all brush single user.
In conclusion technical solution provided by embodiment of the disclosure, it can be according to the interaction between user and businessman
Behavioral data establishes related network, which includes the multiple summits of multiple nodes and the above-mentioned multiple nodes of connection, the side
It is attached to the weight for describing the tightness degree of the interbehavior between user and businessman；Pass through preset label propagation algorithm root
The related network is divided into multiple Web Communities according to the weight of above-mentioned multiple summits；According to the modularity of the Web Community and preset
Modularity threshold value determine there are the target network communities of brush single act.The related network pair that can be made up of user and businessman
Brush single act is identified and is detected to a variety of brush single acts occurred in the form of clique, community, and what raising brush singly detected is applicable in
Property, expand detection range.
Fig. 6 is a kind of block diagram for brushing single detection device shown according to an exemplary embodiment, as shown in fig. 6, the device
600 include:
Network establishes module 610, for establishing related network according to the interbehavior data between user and businessman, the pass
Network of networking includes the multiple summits of multiple nodes and the above-mentioned multiple nodes of connection, which is incidentally used to describe between user and businessman
Interbehavior tightness degree weight；
Community's division module 620, for by preset label propagation algorithm according to the weights of above-mentioned multiple summits by the pass
Networking network is divided into multiple Web Communities；
Single determining module 630 is brushed, for determining exist according to the modularity of the Web Community and preset modularity threshold value
The target network community of brush single act.
Fig. 7 is the block diagram that module is established according to a kind of network shown in Fig. 6, as shown in fig. 7, the node be user node or
Businessman's node, the network establish module 610, comprising:
Node acquisition submodule 611, for using the User ID of user as user node, using the Merchant ID of businessman as quotient
Family's node, to obtain multiple node；
Network setting up submodule 612, for according to the interbehavior data, there are the user node of interbehavior and quotient
The side of subsidiary weight is established between family's node, to establish the related network.
Optionally, which includes the interactive operation and every kind of interactive operation pair that the interbehavior includes
The weighting coefficient and frequency of occurrence answered, the network setting up submodule 612, are used for:
There are establishing side between the user node of interbehavior and businessman's node, to determine the multiple summits；
The one or more target interactive operations for including in the corresponding interbehavior in the first side are obtained, which is that this is more
When any bar in；
The summation of the corresponding operation weight of above-mentioned one or more target interactive operations is calculated, which is the target
The product of the frequency of occurrence of interactive operation weighting coefficient corresponding with the target interactive operation；
Using the summation of the operation weight as the weight on first side, with the subsidiary power of each edge in the above-mentioned multiple summits of determination
Weight.
Fig. 8 be according to a kind of block diagram of community's division module shown in Fig. 6, as shown in figure 8, community's division module 620,
Include:
Label adds submodule 621, for adding different labels for each node in the related network；
Tag update submodule 622, for according to multiple neighbor nodes of above-mentioned each node and above-mentioned each node it
Between side weight, by the label relay algorithm operation is updated to the label of above-mentioned each node；
Community divides submodule 623, for will be provided with phase after completing the operation of the update to the label of above-mentioned each node
Node division with label is consolidated network community, to obtain above-mentioned multiple Web Communities.
Optionally, the tag update submodule 622, is used for:
Determine that multiple target base nodes of destination node, the destination node are any node in the related network；
Target base node division by same label subsidiary in above-mentioned multiple target base nodes is a node group, with
Obtain the corresponding multiple node groups of the destination node；
Determine have the group of destination nodes of maximum label weight in above-mentioned multiple node groups, which is the target
The summation of the weight on the side between all nodes in node and each node group；
It is the second label by the first tag update of the destination node, which is any section in the group of destination nodes
The label of point.
Fig. 9 be according to a kind of block diagram of brush list determining module shown in Fig. 6, as shown in figure 9, the brush list determining module 630,
Include:
Node determines submodule 631, for determining the adjacent node of each node in first network community, the adjoining section
Point is the neighbor node that the first network community is not belonging in the neighbor node of above-mentioned each node, which is upper
State any Web Community in multiple Web Communities；
Community's acquisition submodule 632, for obtaining corresponding second Web Community, the first network community, second network
Community includes the adjacent node of all nodes and above-mentioned all nodes in the first network community；
Modularity acquisition submodule 633, for obtain second while second Web Community include it is all while in it is shared
Ratio, the object module degree as the first network community, wherein the node at the second side both ends belongs to the first network
Community；
Brush is single to determine submodule 634, if being greater than the modularity threshold value for the object module degree, determines the first network society
Area is the target network community.
In conclusion technical solution provided by embodiment of the disclosure, it can be according to the interaction between user and businessman
Behavioral data establishes related network, which includes the multiple summits of multiple nodes and the above-mentioned multiple nodes of connection, the side
It is attached to the weight for describing the tightness degree of the interbehavior between user and businessman；Pass through preset label propagation algorithm root
The related network is divided into multiple Web Communities according to the weight of above-mentioned multiple summits；According to the modularity of the Web Community and preset
Modularity threshold value determine there are the target network communities of brush single act.The related network pair that can be made up of user and businessman
Brush single act is identified and is detected to a variety of brush single acts occurred in the form of clique, community, and what raising brush singly detected is applicable in
Property, expand detection range.
Illustratively, Figure 10 is the block diagram of a kind of electronic equipment 1000 shown according to an exemplary embodiment.For example, electronics
Equipment 1000 may be provided as a server.Referring to Fig.1 0, server 1000 includes processor 1001, and quantity can be one
A or multiple and memory 1002, for storing the computer program that can be executed by processor 1001.It is deposited in memory 1002
The computer program of storage may include it is one or more each correspond to one group of instruction module.In addition, processor
1001 can be configured as the execution computer program, to execute the speed adjustment method of above-mentioned Data Migration.
In addition, server 1000 can also include power supply module 1003 and communication component 1004, which can
To be configured as the power management of execute server 1000, which, which can be configured as, realizes server 1000
Communication, for example, wired or wireless communication.In addition, the server 1000 can also include input/output (I/O) interface 1005.Clothes
Business device 1000 can be operated based on the operating system for being stored in memory 1002, such as Windows ServerTM, Mac OS
XTM, UnixTM, LinuxTM etc..
In a further exemplary embodiment, a kind of computer readable storage medium including program instruction is additionally provided, it should
The step of speed adjustment method of above-mentioned Data Migration is realized when program instruction is executed by processor.For example, the computer can
Reading storage medium can be the above-mentioned memory 1002 including program instruction, and above procedure instruction can be by the processing of server 1000
Device 1001 is executed to complete the speed adjustment method of above-mentioned Data Migration.
The preferred embodiment of the disclosure is described in detail in conjunction with attached drawing above, still, the disclosure is not limited to above-mentioned reality
The detail in mode is applied, in the range of the technology design of the disclosure, a variety of letters can be carried out to the technical solution of the disclosure
Monotropic type, these simple variants belong to the protection scope of the disclosure.
It is further to note that specific technical features described in the above specific embodiments, in not lance
In the case where shield, can be combined in any appropriate way, in order to avoid unnecessary repetition, the disclosure to it is various can
No further explanation will be given for the combination of energy.
Claims (10)
1. a kind of single detection method of brush, which is characterized in that the described method includes:
Establish related network according to the interbehavior data between user and businessman, the related network include multiple nodes and
The multiple summits of the multiple node are connected, the side is incidentally used to describe the tightness degree of the interbehavior between user and businessman
Weight；
The related network is divided by multiple network societies according to the weight of the multiple summits by preset label propagation algorithm
Area；
Determine that there are the target network communities of brush single act according to the modularity of the Web Community and preset modularity threshold value.
2. the method according to claim 1, wherein the interbehavior data according between user and businessman
Establish related network, comprising:
Using the User ID of the user as user node, using the Merchant ID of the businessman as businessman's node, described in obtaining
Multiple nodes；
According to the interbehavior data, there are establish subsidiary weight between the user node of interbehavior and businessman's node
Side, to establish the related network.
3. according to the method described in claim 2, it is characterized in that, the interbehavior data include that the interbehavior includes
Interactive operation and the corresponding weighting coefficient of every kind of interactive operation and frequency of occurrence, it is described according to the interbehavior data,
There are the sides for establishing subsidiary weight between the user node of interbehavior and businessman's node, to establish the related network, packet
It includes:
There are side is established between the user node of interbehavior and businessman's node, with the determination multiple summits；
The one or more target interactive operations for including in the corresponding interbehavior in the first side are obtained, first side is described more
When any bar in；
The summation of the corresponding operation weight of one or more target interactive operations is calculated, the operation weight is the target
The product of the frequency of occurrence of interactive operation weighting coefficient corresponding with the target interactive operation；
Using the summation of the operation weight as the weight on first side, with the subsidiary power of each edge in the determination multiple summits
Weight.
4. according to the method described in claim 2, it is characterized in that, it is described by preset label propagation algorithm according to described more
The related network is divided into multiple Web Communities by the weight on side, comprising:
Different labels is added for each node in the related network；
According to the weight on the side between each node and multiple neighbor nodes of each node, turned by the label
It broadcasts algorithm and operation is updated to the label of each node；
After completing the operation of the update to the label of each node, the node division that will be provided with same label is consolidated network
Community, to obtain the multiple Web Community.
5. according to the method described in claim 4, it is characterized in that, described according to each node and each node
The weight on the side between multiple neighbor nodes is updated operation to the label of each node, comprising:
Determine that multiple target base nodes of destination node, the destination node are any node in the related network；
Target base node division by same label subsidiary in the multiple target base node is a node group, to obtain
The corresponding multiple node groups of the destination node；
Determine have the group of destination nodes of maximum label weight in the multiple node group, the label weight is the target
The summation of the weight on the side between all nodes in node and each node group；
It is the second label by the first tag update of the destination node, second label is any in the group of destination nodes
The label of node.
6. the method according to claim 1, wherein the modularity according to the Web Community and preset
Modularity threshold value determines that there are the target network communities of brush single act, comprising:
The adjacent node of each node in first network community is obtained, the adjacent node is that the neighbours of each node save
The neighbor node of the first network community is not belonging in point, the first network community is appointing in the multiple Web Community
One Web Community；
Corresponding second Web Community, the first network community is obtained, second Web Community includes the first network society
The adjacent node of all nodes and all nodes in area；
Obtain second while second Web Community include it is all while in shared ratio, as the first network community
Object module degree, wherein the node at second side both ends belongs to the first network community；
If the object module degree is greater than the modularity threshold value, determine that the first network community is the target network society
Area.
7. a kind of single detection device of brush, which is characterized in that described device includes:
Network establishes module, for establishing related network, the association net according to the interbehavior data between user and businessman
Network includes the multiple summits of multiple nodes and the multiple node of connection, and the side is incidentally used to describe between user and businessman
The weight of the tightness degree of interbehavior；
Community's division module, for by preset label propagation algorithm according to the weights of the multiple summits by the related network
It is divided into multiple Web Communities；
Single determining module is brushed, for determining that there are brush unilines according to the modularity of the Web Community and preset modularity threshold value
For target network community.
8. device according to claim 7, which is characterized in that the network establishes module, comprising:
Node acquisition submodule, for using the User ID of the user as user node, using the Merchant ID of the businessman as
Businessman's node, to obtain the multiple node；
Network setting up submodule, for according to the interbehavior data, there are the user nodes of interbehavior and businessman's section
The side of subsidiary weight is established between point, to establish the related network.
9. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is held by processor
The step of brush list detection method described in any one of claims 1 to 6 is realized when row.
10. a kind of electronic equipment characterized by comprising
Memory is stored thereon with computer program；
Processor, for executing the computer program in the memory, to realize any one of claims 1 to 6 institute
The step of stating method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910069412.3A CN109598563A (en) | 2019-01-24 | 2019-01-24 | Brush single detection method, device, storage medium and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910069412.3A CN109598563A (en) | 2019-01-24 | 2019-01-24 | Brush single detection method, device, storage medium and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109598563A true CN109598563A (en) | 2019-04-09 |
Family
ID=65964749
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910069412.3A Pending CN109598563A (en) | 2019-01-24 | 2019-01-24 | Brush single detection method, device, storage medium and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109598563A (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110610365A (en) * | 2019-09-17 | 2019-12-24 | 中国建设银行股份有限公司 | Method and device for identifying transaction request |
| CN110633994A (en) * | 2019-07-12 | 2019-12-31 | 中国农业银行股份有限公司 | Identification method and device for single swiping behavior |
| CN111599472A (en) * | 2020-05-14 | 2020-08-28 | 重庆大学 | Method and device for recognizing psychological states of students and computer |
| CN112001649A (en) * | 2020-08-27 | 2020-11-27 | 支付宝(杭州)信息技术有限公司 | Risk data mining method, device and equipment |
| CN112052404A (en) * | 2020-09-23 | 2020-12-08 | 西安交通大学 | Group discovery method, system, device and medium for multi-source heterogeneous relation network |
| CN112184267A (en) * | 2020-10-27 | 2021-01-05 | 北京嘀嘀无限科技发展有限公司 | Method, apparatus, device and medium for discovering user group in service application |
| CN112184334A (en) * | 2020-10-27 | 2021-01-05 | 北京嘀嘀无限科技发展有限公司 | Method, apparatus, device and medium for determining problem users |
| CN112288528A (en) * | 2020-10-30 | 2021-01-29 | 浙江集享电子商务有限公司 | Malicious community discovery method and device, computer equipment and readable storage medium |
| CN112734506A (en) * | 2019-10-14 | 2021-04-30 | 阿里巴巴集团控股有限公司 | Data searching method, data detecting method, device and equipment |
| CN112837078A (en) * | 2021-03-03 | 2021-05-25 | 万商云集(成都)科技股份有限公司 | Cluster-based user abnormal behavior detection method |
| CN112907308A (en) * | 2019-11-19 | 2021-06-04 | 京东数字科技控股有限公司 | Data detection method and device and computer readable storage medium |
| CN113157767A (en) * | 2021-03-24 | 2021-07-23 | 支付宝(杭州)信息技术有限公司 | Risk data monitoring method, device and equipment |
| CN113724054A (en) * | 2021-09-10 | 2021-11-30 | 中国银行股份有限公司 | Method and device for detecting manual bill brushing |
| CN113763077A (en) * | 2020-07-24 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | Method and apparatus for detecting false trade orders |
| CN114338216A (en) * | 2021-12-31 | 2022-04-12 | 招商银行股份有限公司 | Multidimensional brute-force attack prevention method, apparatus, device, medium, and program product |
-
2019
- 2019-01-24 CN CN201910069412.3A patent/CN109598563A/en active Pending
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110633994A (en) * | 2019-07-12 | 2019-12-31 | 中国农业银行股份有限公司 | Identification method and device for single swiping behavior |
| CN110610365A (en) * | 2019-09-17 | 2019-12-24 | 中国建设银行股份有限公司 | Method and device for identifying transaction request |
| CN112734506A (en) * | 2019-10-14 | 2021-04-30 | 阿里巴巴集团控股有限公司 | Data searching method, data detecting method, device and equipment |
| CN112907308B (en) * | 2019-11-19 | 2024-05-24 | 京东科技控股股份有限公司 | Data detection method and device, and computer readable storage medium |
| CN112907308A (en) * | 2019-11-19 | 2021-06-04 | 京东数字科技控股有限公司 | Data detection method and device and computer readable storage medium |
| CN111599472A (en) * | 2020-05-14 | 2020-08-28 | 重庆大学 | Method and device for recognizing psychological states of students and computer |
| CN111599472B (en) * | 2020-05-14 | 2023-10-24 | 重庆大学 | Method and device for identifying psychological state of student and computer |
| CN113763077A (en) * | 2020-07-24 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | Method and apparatus for detecting false trade orders |
| WO2022017082A1 (en) * | 2020-07-24 | 2022-01-27 | 北京沃东天骏信息技术有限公司 | Method and apparatus for detecting false transaction orders |
| CN112001649A (en) * | 2020-08-27 | 2020-11-27 | 支付宝(杭州)信息技术有限公司 | Risk data mining method, device and equipment |
| CN112052404A (en) * | 2020-09-23 | 2020-12-08 | 西安交通大学 | Group discovery method, system, device and medium for multi-source heterogeneous relation network |
| CN112052404B (en) * | 2020-09-23 | 2023-08-15 | 西安交通大学 | Group discovery method, system, equipment and medium of multi-source heterogeneous relation network |
| CN112184267A (en) * | 2020-10-27 | 2021-01-05 | 北京嘀嘀无限科技发展有限公司 | Method, apparatus, device and medium for discovering user group in service application |
| CN112184334A (en) * | 2020-10-27 | 2021-01-05 | 北京嘀嘀无限科技发展有限公司 | Method, apparatus, device and medium for determining problem users |
| CN112288528A (en) * | 2020-10-30 | 2021-01-29 | 浙江集享电子商务有限公司 | Malicious community discovery method and device, computer equipment and readable storage medium |
| CN112837078A (en) * | 2021-03-03 | 2021-05-25 | 万商云集(成都)科技股份有限公司 | Cluster-based user abnormal behavior detection method |
| CN112837078B (en) * | 2021-03-03 | 2023-11-03 | 万商云集(成都)科技股份有限公司 | Method for detecting abnormal behavior of user based on clusters |
| CN113157767A (en) * | 2021-03-24 | 2021-07-23 | 支付宝(杭州)信息技术有限公司 | Risk data monitoring method, device and equipment |
| CN113157767B (en) * | 2021-03-24 | 2022-06-07 | 支付宝(杭州)信息技术有限公司 | Risk data monitoring method, device and equipment |
| CN113724054A (en) * | 2021-09-10 | 2021-11-30 | 中国银行股份有限公司 | Method and device for detecting manual bill brushing |
| CN114338216A (en) * | 2021-12-31 | 2022-04-12 | 招商银行股份有限公司 | Multidimensional brute-force attack prevention method, apparatus, device, medium, and program product |
| CN114338216B (en) * | 2021-12-31 | 2024-03-26 | 招商银行股份有限公司 | Multidimensional brushing attack prevention and control method, device, equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109598563A (en) | Brush single detection method, device, storage medium and electronic equipment | |
| CN106126521B (en) | The social account method for digging and server of target object | |
| Ju et al. | A new algorithm for positive influence maximization in signed networks | |
| Pinna et al. | A petri nets model for blockchain analysis | |
| Starnini et al. | Random walks on temporal networks | |
| CN103678669B (en) | Evaluating system and method for community influence in social network | |
| Ribeiro et al. | Variable neighborhood search for the degree-constrained minimum spanning tree problem | |
| Victoire et al. | A modified hybrid EP–SQP approach for dynamic dispatch with valve-point effect | |
| Li et al. | Labeled influence maximization in social networks for target marketing | |
| Sela et al. | Active viral marketing: Incorporating continuous active seeding efforts into the diffusion model | |
| Zheng et al. | Improving the efficiency of multi-objective evolutionary algorithms through decomposition: An application to water distribution network design | |
| CN109829337A (en) | A kind of method, system and the equipment of community network secret protection | |
| CN107895038A (en) | A kind of link prediction relation recommends method and device | |
| CN110009365B (en) | User group detection method, device and equipment for abnormally transferring electronic assets | |
| Yeshwanth et al. | Evolutionary churn prediction in mobile networks using hybrid learning | |
| CN105913235A (en) | Client account transfer relation analysis method and system | |
| CN112566093A (en) | Terminal relation identification method and device, computer equipment and storage medium | |
| Hu et al. | A new algorithm CNM-Centrality of detecting communities based on node centrality | |
| CN109978575A (en) | A kind of method and device excavated customer flow and manage scene | |
| Chen et al. | Influential node detection of social networks based on network invulnerability | |
| Wang et al. | A two phase removing algorithm for minimum independent dominating set problem | |
| CN107507020B (en) | Method for obtaining network propagation influence competitive advantage maximization | |
| CN109802859A (en) | Nodes recommendations method and server in a kind of network | |
| Tejaswi et al. | Target specific influence maximization: An approach to maximize adoption in labeled social networks | |
| CN111666501A (en) | Abnormal community identification method and device, computer 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 |