WO2019100967A1 - 用于识别异常交易社团的方法和装置 - Google Patents

用于识别异常交易社团的方法和装置 Download PDF

Info

Publication number
WO2019100967A1
WO2019100967A1 PCT/CN2018/115141 CN2018115141W WO2019100967A1 WO 2019100967 A1 WO2019100967 A1 WO 2019100967A1 CN 2018115141 W CN2018115141 W CN 2018115141W WO 2019100967 A1 WO2019100967 A1 WO 2019100967A1
Authority
WO
WIPO (PCT)
Prior art keywords
transaction
community
node
nodes
abnormal
Prior art date
Application number
PCT/CN2018/115141
Other languages
English (en)
French (fr)
Inventor
李旭瑞
郑建宾
赵金涛
Original Assignee
中国银联股份有限公司
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 中国银联股份有限公司 filed Critical 中国银联股份有限公司
Publication of WO2019100967A1 publication Critical patent/WO2019100967A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/901Indexing; Data structures therefor; Storage structures
    • G06F16/9024Graphs; Linked lists
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q50/00Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/01Social networking

Definitions

  • the present invention relates to data processing techniques, and more particularly to a method for identifying an abnormal transaction community, an apparatus for implementing the method, and a computer readable storage medium comprising the computer program implementing the method.
  • An object of the present invention is to provide a method for identifying an abnormal transaction community, which has the advantages of high processing efficiency, high recognition accuracy, and the like.
  • a method for identifying an abnormal transaction community in accordance with an aspect of the present invention includes the following steps:
  • each node of the network map represents one of the plurality of accounts, and is indicated by two sides connecting the two nodes A transaction occurs between the accounts associated with the node, where the direction of the side represents the direction of the transaction;
  • the corresponding risk metric is determined according to the transaction information of the community, and the risk metric is used to determine whether the community belongs to the abnormal trading community.
  • the step of determining a community comprises:
  • the division operation is performed in the following manner:
  • the connected subgraph is subjected to community partitioning in an iterative manner until the modularity of the connected subgraph does not change after the partitioning, thereby completing the community partitioning of the connected subgraph.
  • the node weight depends on the transaction amount, the number of transactions, and the total number of penalties of each node at both ends of the edge.
  • the transaction timing depends on the average transfer time of funds and the average transfer time of funds for each node at both ends of the edge.
  • the contribution value to the module degree is related to the direction of the edge.
  • the transaction information includes the time of each transaction within each community, the total number of transactions of the community, and the total transaction amount.
  • the risk metric of each community includes the transaction time entropy of the association and the overall risk factor.
  • Still another object of the present invention is to provide an apparatus for identifying an abnormal transaction community, which has the advantages of high processing efficiency, high recognition accuracy, and the like.
  • An apparatus for identifying an abnormal transaction community includes:
  • a first module configured to construct a network map related to transaction events between the plurality of accounts, wherein each node of the network map represents one of the plurality of accounts, and the side connecting the two nodes Indicating that a transaction has occurred between accounts associated with the two nodes, where the direction of the side represents the direction of the transaction;
  • a second module configured to determine from the network map as one or more communities
  • the third module is configured to determine a corresponding risk metric according to the transaction information of the community, and the risk metric is used to determine whether the community belongs to the abnormal transaction community.
  • An apparatus for identifying an abnormal transaction community in accordance with another aspect of the present invention includes a memory, a processor, and a computer program stored on the memory and operative on the processor to perform the method as described above.
  • FIG. 1 is a flow chart of a method for identifying an abnormal transaction community in accordance with one embodiment of the present invention.
  • FIG. 2 is a flow chart of a method of determining a community applicable to the embodiment of FIG. 1.
  • FIG. 3 is a flow chart of a community partitioning algorithm applicable to the embodiment of FIG. 2.
  • FIG. 4 is a flow chart of an iterative algorithm applicable to the embodiment of FIG.
  • FIG. 5 is a flow diagram of a method of determining a risk metric for a community that may be applied to the embodiment of FIG. 1.
  • FIG. 6 is a block diagram of an apparatus for identifying an abnormal transaction community in accordance with another embodiment of the present invention.
  • FIG. 7 is a block diagram of an apparatus for identifying an abnormal transaction community in accordance with another embodiment of the present invention.
  • FIG. 1 is a flow chart of a method for identifying an abnormal transaction community in accordance with one embodiment of the present invention.
  • the method illustrated in Figure 1 can be performed at a cloud server or a background transaction processing system.
  • the flow of the method illustrated in FIG. 1 begins at step 110.
  • selecting a time period between a plurality of transaction accounts within the T m characterize and construct a plurality of transaction accounts FIG networks between each event.
  • the network map can be constructed, for example, in such a way that each node of the network map represents one of a plurality of accounts, and the side connecting the two nodes indicates that a transaction has occurred between the accounts associated with the two nodes.
  • the edge is a directed edge whose direction indicates the direction of the transaction (for example, in a transaction, the direction may be defined as a transfer node pointing from the transfer node of the fund to the fund, but defining it as The transfer node of funds points to the transfer node of funds is equivalent).
  • the sides have weights.
  • the initial weight W Bi of the i-th edge in the network map can be set to:
  • ⁇ m and ⁇ c are the coefficients corresponding to the total transaction amount and the total transaction number, respectively.
  • the sum of the coefficients is 1.
  • the network map generated from step 110 is determined to be one or more communities. The specific ways in which the association is determined will be described in detail below.
  • a corresponding risk metric is determined according to the transaction information, and the risk metric is used to determine whether the community has an abnormal transaction community. Specific ways to determine risk metrics are described in more detail below.
  • FIG. 2 is a flow chart of a method of determining a community applicable to the embodiment of FIG. 1.
  • the method illustrated in Figure 2 can be performed at a cloud server or a background transaction processing system.
  • one or more connected sub-pictures are determined from the network map generated at step 110.
  • the determining sub-graph is determined by first filtering out isolated nodes in the network graph (ie, nodes that have no transactions with other nodes), and then dividing the entire network graph into one or more connected sub-graphs (eg, utilizing The connected component algorithm is such that, within each of the divided sub-graphs, any two nodes are connected, and there is no connected edge between the two connected sub-pictures.
  • a subset is selected from the connected sub-pictures determined in step 210.
  • the elements in the subset can be selected in the following manner: First, select the total number of nodes in the medium-sized connected subgraph. Then, in these medium-sized connected subgraphs, the amount of transfer and/or the number of transactions (hereinafter referred to as “outside”) or the amount of transfer and/or the number of transactions transferred (hereinafter referred to as “input degree” are counted. ”) The number of larger nodes, collectively referred to as suspicious central nodes. Finally, the connected subgraphs with a large number of suspicious central nodes in these medium-sized connected subgraphs are selected into the subset.
  • a node whose degree of ingress (indegree) is greater than a threshold may be regarded as a suspicious central node, and the threshold may be set, for example, to generate an outdegree (into degree) of all nodes in the connected subgraph.
  • the distribution map is statistically set, and the curve turning point in the statistical distribution map is set as the threshold of the outdegree (into the degree).
  • a connected subgraph having a number of suspected central nodes greater than a threshold may also be included in the subset.
  • step 230 a community partitioning operation is performed on the connected sub-graphs in the subset that have not been subjected to the community partitioning operation.
  • a detailed description of the community division operation will be given below.
  • step 240 it is determined whether a community partitioning operation is performed for each of the connected sub-graphs in the subset, and if so, step 130 of FIG. 1 may be entered, otherwise returning to step 230.
  • steps 210 and 220 are preferred steps. That is, in one embodiment, the community partitioning operation as described below may be performed directly on the network map, or the community partitioning operation may be performed on each of the connected sub-graphs determined in step 210.
  • Each connected subgraph can be considered as a trading group with an associated nature.
  • unusual trading activities such as money laundering.
  • some executors of illegal trading activities will deliberately hide the core abnormal trading structure in a large number of normal transactions, which further increases the difficulty of finding abnormal trading.
  • the inventors of the present invention have found through in-depth research that if a connected subgraph is directly analyzed or collated, it is likely that the result is that although the risk measure for the abnormal transaction for measuring the connected subgraph is low, actually There are a lot of unusual transactions hidden.
  • the inventors of the present invention creatively introduce the following ways to mine hidden abnormal transactions: the weights of the edges in the connected subgraph are corrected based on the node weight and the transaction timing, and then the modularity specifically defined for the directed graph is utilized.
  • the iteratively, the connected subgraphs of the edges are subjected to community division, until the modularity of the connected subgraphs does not change after the partitioning, thereby completing the community partitioning of the connected subgraphs.
  • FIG. 3 is a flow chart of a community partitioning algorithm applicable to the embodiment of FIG. 2, which is based on the above manner.
  • the operation object of the algorithm shown in Fig. 3 is a connected subgraph, but this is merely exemplary, and it is also true that the entire network map is regarded as an operation object.
  • the flow shown in FIG. 3 begins at step 310.
  • the weight of each edge of a connected subgraph is corrected or optimized using the node weight.
  • the node weights for correcting the weight of the edge may be calculated using transaction information such as the transaction amount, the number of transactions, and the total number of degrees of entry of one node.
  • the specific calculation method is as shown in the following formula (2):
  • ⁇ vj is the node weight of node j
  • the normalized value of the total transaction amount of the node j, the normalized value of the transaction number, and the normalized value of the total number of the entry degrees, ⁇ Mv , ⁇ Cv , ⁇ Dv are the total transaction amount of the node j, the number of transactions, and the total weight of the degree of entry and exit.
  • the factor (for example, each weighting factor can take a value of 1/3).
  • the weights of the edges (2) and (3) can be used to correct the weights, so that the weights of the edges are corrected by the node weights.
  • step 320 the transaction timing correction or optimization is further performed on the weight of the edge of the connected subgraph corrected by the node weight.
  • the transaction timing correction or optimization is further performed on the weight of the edge of the connected subgraph corrected by the node weight.
  • further modifications can be made in the following manner.
  • the average connection time of node A is:
  • the weight correction factor associated with the transaction timing is then determined.
  • first decentralized transfer and then transferred out that is, first, the transfer process in which multiple nodes transfer money to one node, and then the latter transfers the aggregated amount in a centralized manner
  • focus on The edge that is produced should be formed after the edges that have been dispersed many times.
  • distributed and transfer after centralized transfer that is, first, a node receives a payment, and then the node transfers the money to multiple nodes, and finally multiple nodes transfer the money received by each node.
  • the transaction process from the transaction timing, the side that is transferred intensively should be formed before the scattered edges are scattered.
  • weight correction coefficient ⁇ 1 is determined according to the following formula:
  • T src is the time when the initial node src is connected to the j-th edge
  • T R is the normalization factor
  • the corresponding weight correction coefficient ⁇ 2 is determined according to the following formula:
  • T dst is the time when the destination node dst is connected to the j-th edge
  • T R is the normalization factor
  • W Ei is the weight of the used node weight determined by the i-th edge determined in step 310.
  • step 330 in which the connected sub-pictures subjected to the weight correction processing of steps 310 and 320 are subjected to community division, thereby classifying each node into the corresponding community.
  • each side is a directed side.
  • i ⁇ j Represents the weight sum of all edges pointing to node i
  • k i represents the weight sum of all edges of node j
  • k j represents the weight sum of all edges of node j.
  • the module degree Q D can be defined as:
  • an iterative algorithm similar to the Louvain algorithm can be employed to perform community partitioning using the modularity defined above.
  • FIG. 4 is a flow chart of an iterative algorithm applicable to the embodiment of FIG.
  • step 410 an initialization process is first performed to classify each node in a connected subgraph into a different community.
  • step 420 an iterative operation is performed for each node in the connected subgraph using the modularity defined by the above equation (15).
  • the node i is first assigned to the community to which each of its neighbor nodes belongs, and then the module change value before and after the allocation is calculated, thereby obtaining the association with the node i.
  • One or more module degree change values can be determined according to the following formula:
  • step 430 it is determined whether the state of the belonging community of all nodes changes before and after the current execution step 420. If the change occurs, the process returns to step 420, otherwise, the process proceeds to step 440.
  • step 440 the connected subgraph is compressed in the following manner: the nodes belonging to the same community are compressed into a new node, the weights of the edges between the nodes in the community are converted into the weights of the rings of the new node, and the edge weights between the communities are transformed. The edge weight between the new nodes.
  • step 450 the modularity of the compressed connected sub-picture generated in step 440 is determined in accordance with the above equations (15)-(17), and then proceeds to step 460.
  • step 460 it is determined whether the difference between the module degree determined in step 450 and the modularity of the connected sub-picture before the step 440 is less than a preset threshold. If yes, the process proceeds to step 470, where the currently processed connected sub-picture is output. The result of the community division, otherwise returns to step 420.
  • FIG. 5 is a flow diagram of a method of determining a risk metric for a community that may be applied to the embodiment of FIG. 1. For the sake of convenience, the description herein determines an example of a risk metric for a community k.
  • the flow shown in FIG. 5 begins at step 510.
  • determining the average trading time of the society for which the risk measure is to be determined during the time period T m Preferably, for each transaction of the association during the period of time, the transaction time can be determined by using the first transaction as the time reference point.
  • step 520 For each transaction of the association during that period of time, determine the absolute value ⁇ T h of the difference between its trading time and the average trading time, where h is the index number of the transaction.
  • each transaction is classified into corresponding intervals of a plurality of intervals according to the value of ⁇ T h , and the ratio of the number of transactions in each interval to the total number of transactions of the society during the time period T m is counted. .
  • a transaction time entropy H C for reflecting the correlation between the transaction time and the abnormal transaction is determined according to the following formula:
  • n is the total number of intervals
  • P i represents the ratio of the number of transactions in the i-th interval to the total number of transactions of the society during the time period T m .
  • an overall risk factor for the community is determined.
  • the overall risk factor ⁇ k can be determined using the following equation:
  • a standardized value for the number of nodes within the community k The normalized value of the total number of transactions for the community k during the time period T m , The normalized value of the total transaction amount for the community k during the time period T m , The normalized value of the average degree of the nodes in the community k, a normalized value of the transaction time entropy for the community k during the time period T m , It is a weight value and can be set according to the actual application.
  • a plurality of communities in a network diagram or a connected subgraph they can be ranked from highest to lowest according to the overall risk factor determined by the method shown in FIG. 5, wherein the top 5%
  • the club is rated as a level I suspicious group, and between 5% and 10% of the clubs are rated as level II suspicious societies.
  • FIG. 1-5 describes a method for identifying a time period in association unusual transactions in T m.
  • the above embodiments can also be extended to the identification of abnormal transaction communities within a plurality of time periods. When it is desired to monitor trading activity over a longer span of time periods, it is advantageous to divide the long span time period into multiple time periods in view of possible changes in the community.
  • a longer span of time periods (eg, one week, one month, or half a year, etc.) can be divided into n time periods, and then in each time period, the embodiments described above with reference to FIGS. 1-5 are employed, respectively.
  • Identify unusual trading communities In view of the large amount of data, it is preferable to divide the community by the following incremental method. Specifically, after the community division is completed in the first time period T i , the community label corresponding to each node is retained; and then, when the community is divided into the next time period T i+1 , the time period is taken.
  • intersection of all nodes with all the nodes in the previous time period, and the community label corresponding to the node of the intersection part is used as the initial label of the relevant node of the current time period, and the nodes without the social label are initialized as the association to which they belong. Then, the community division operation is performed on this basis. This method can greatly speed up the convergence speed of community division operations.
  • FIG. 6 is a block diagram of an apparatus for identifying an abnormal transaction community in accordance with another embodiment of the present invention.
  • the apparatus 60 shown in FIG. 6 includes a memory 610, a processor 620, and a computer program 630 stored on the memory 610 and executable on the processor 620, wherein the computer program 630 is executable by operating on the processor 620
  • FIG. 7 is a block diagram of an apparatus for identifying an abnormal transaction community in accordance with another embodiment of the present invention.
  • the apparatus 70 shown in FIG. 7 includes a first module 710, a second module 720, and a third module 730, wherein the first module 710 is configured to construct a network map related to transaction events between a plurality of accounts, wherein Each node of the network map represents one of the plurality of accounts, and an edge connecting the two nodes indicates that a transaction has occurred between the accounts associated with the two nodes, wherein the direction of the side represents the direction of the transaction a second module 720 for determining from the network map as one or more communities; and a third module 730 for determining a corresponding risk metric according to the transaction information of the community, the risk metric being used to determine whether the community is abnormal Trading community.
  • a computer readable storage medium having stored thereon a computer program that, when executed by a processor, implements the method of the embodiment described with reference to Figures 1-3.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Tourism & Hospitality (AREA)
  • Economics (AREA)
  • Primary Health Care (AREA)
  • Strategic Management (AREA)
  • Human Resources & Organizations (AREA)
  • General Health & Medical Sciences (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Health & Medical Sciences (AREA)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

一种数据处理技术,特别涉及用于识别异常交易社团方法、实施该方法的装置(60)以及包含实施该方法的计算机程序的计算机可读存储介质。所述用于识别异常交易社团的方法包含下列步骤:构建与多个账户相互间的交易事件相关的网络图,其中,所述网络图的每个节点代表所述多个账户的其中一个,并且以连接两个节点的边来指示在与这两个节点相关联的账户之间发生了交易,其中边的方向代表交易的方向;从所述网络图确定为一个或多个社团;以及根据社团的交易信息确定其相应的风险度量,该风险度量用于确定该社团是否属于异常交易社团。

Description

用于识别异常交易社团的方法和装置 技术领域
本发明涉及数据处理技术,特别涉及用于识别异常交易社团方法、实施该方法的装置以及包含实施该方法的计算机程序的计算机可读存储介质。
背景技术
诸如洗钱之类的非法资金转移由于其对国家金融体系安全和经济秩序稳定带来的危害,一直是政府监管的重点。随着电子支付的兴起,更加便捷的支付方式在提高交易效率和降低交易成本的同时,也给非法资金转移提供了可乘之机。
目前主流的反洗钱(AML)系统大多是基于规则的。这类系统的缺点是监管效率较低,并且由于规则很容易被学习掌握,导致监管被规避。此外,规则系统包含较多的主观因素,难免出现错误或者疏漏。再者,由于洗钱之类的资金非法转移活动往往涉及团伙犯罪,当前的监管系统缺乏全局性的监测能力,从而难以发现大范围内的洗钱活动。
有鉴于此,迫切需要一种能够准确、快速地识别异常交易社团的方法和装置。
发明内容
本发明的一个目的是提供一种用于识别异常交易社团的方法,其具有处理效率高、识别准确度高等优点。
按照本发明一个方面的用于识别异常交易社团的方法包含下列步骤:
构建与多个账户相互间的交易事件相关的网络图,其中,所述网络图的每个节点代表所述多个账户的其中一个,并且以连接两个节点的边来指示在与这两个节点相关联的账户之间发生了交易,其中边的 方向代表交易的方向;
从所述网络图确定为一个或多个社团;以及
根据社团的交易信息确定其相应的风险度量,该风险度量用于确定该社团是否属于异常交易社团。
优选地,在上述方法中,确定社团的步骤包括:
从所述网络图确定一个或多个连通子图,其中,每个连通子图内的任意两个节点之间是连通的,并且两个连通子图之间无相连接的边;以及
对连通子图执行社团划分操作。
优选地,在上述方法中,在执行社团划分的步骤中,对于任一连通子图,按照下列方式执行划分操作:
基于节点权重和交易时序,对该连通子图中的边的权重进行修正;以及
以迭代方式对该连通子图进行社团划分直到划分后该连通子图的模块度不再变化为止,由此完成该连通子图的社团划分。
优选地,在上述方法中,节点权重依赖于边两端的每个节点的交易金额、交易次数和出入度总数。
优选地,在上述方法中,交易时序依赖于边两端的每个节点的资金平均转入时间和资金平均转出时间。
优选地,在上述方法中,对于两个节点之间的边,其对模块度的贡献值与边的方向相关。
优选地,在上述方法中,所述交易信息包括每个社团内的每笔交易的时间、该社团的总交易数量和总交易金额。
优选地,在上述方法中,每个社团的风险度量包括该社团的交易时间熵和整体风险因子。
本发明的还有一个目的是提供一种用于识别异常交易社团的装置,其具有处理效率高、识别准确度高等优点。
按照本发明另一个方面的用于识别异常交易社团的装置包含:
第一模块,用于构建与多个账户相互间的交易事件相关的网络图,其中,所述网络图的每个节点代表所述多个账户的其中一个,并且以连接两个节点的边来指示在与这两个节点相关联的账户之间发生了交易,其中边的方向代表交易的方向;
第二模块,用于从所述网络图确定为一个或多个社团;以及
第三模块,用于根据社团的交易信息确定其相应的风险度量,该风险度量用于确定该社团是否属于异常交易社团。
按照本发明另一个方面的用于识别异常交易社团的装置包含存储器、处理器以及存储在所述存储器上并可在所述处理器上运行的计算机程序以执行如上所述的方法。
本发明的还有一个目的是提供一种计算机可读存储介质,其上存储计算机程序,该程序被处理器执行时实现如上所述的方法。
附图说明
本发明的上述和/或其它方面和优点将通过以下结合附图的各个方面的描述变得更加清晰和更容易理解,附图中相同或相似的单元采用相同的标号表示。附图包括:
图1为按照本发明一个实施例的用于识别异常交易社团的方法的流程图。
图2为可应用于图1所示实施例的确定社团方法的流程图。
图3为可应用于图2所示实施例的社团划分算法的流程图。
图4为可应用于图3所示实施例的迭代算法的流程图。
图5为可应用于图1所示实施例的确定社团的风险量度的方法的流程图。
图6为按照本发明另一个实施例的用于识别异常交易社团的装置的框图。
图7为按照本发明另一个实施例的用于识别异常交易社团的装置的框图。
具体实施方式
下面参照其中图示了本发明示意性实施例的附图更为全面地说明本发明。但本发明可以按不同形式来实现,而不应解读为仅限于本文给出的各实施例。给出的上述各实施例旨在使本文的披露全面完整,以将本发明的保护范围更为全面地传达给本领域技术人员。
在本说明书中,诸如“包含”和“包括”之类的用语表示除了具有在说明书和权利要求书中有直接和明确表述的单元和步骤以外,本发明的技术方案也不排除具有未被直接或明确表述的其它单元和步骤的情形。
图1为按照本发明一个实施例的用于识别异常交易社团的方法的流程图。优选地但非必须地,图1所示的方法可在云端服务器或后台交易处理系统处执行。
图1所示的方法的流程开始于步骤110。在该步骤中,选取一个时间段T m内的多个账户之间的交易记录,并构建刻画多个账户相互间的交易事件的网络图。该网络图例如可以按照下列方式构建:网络图的每个节点代表多个账户的其中一个,并且以连接两个节点的边来指示在与这两个节点相关联的账户之间发生了交易。在本实施例中,边为有向边,其方向表示交易的方向(例如在一笔交易中,该方向可以定义为从资金的转出节点指向资金的转入节点,但是将其定义为从资金的转入节点指向资金的转出节点是等价的)。此外,在本实施例中,边具有权重。示例性地,可以将网络图中的第i条边的初始权重W Bi设定为:
Figure PCTCN2018115141-appb-000001
这里
Figure PCTCN2018115141-appb-000002
Figure PCTCN2018115141-appb-000003
分别代表边(也即边两端节点之间)的总交易金额的标准化值和总交易次数的标准化值,ω m和ω c分别为总交易金额和总交易 次数所对应的系数,这两个系数之和为1。
随后进入步骤120,从步骤110生成的网络图确定为一个或多个社团。有关社团确定的具体方式将在下面作详细的描述。
接着进入步骤130,对于每个社团,根据其交易信息确定相应的风险度量,该风险度量用于确定该社团是否发生异常交易社团。有关确定风险量度的具体方式将在下面作详细的描述。
图2为可应用于图1所示实施例的确定社团方法的流程图。优选地但非必须地,图2所示的方法可在云端服务器或后台交易处理系统处执行。
如图2所示,在步骤210,从步骤110生成的网络图确定一个或多个连通子图。示例性地,连通子图的确定过程为,首先滤除网络图中的孤立节点(也即与其它节点无交易的节点),然后将整个网络图划分为一个或多个连通子图(例如利用连通分量算法),使得在划分后的每个连通子图内,任意两个节点之间是连通的,并且两个连通子图之间无相连接的边。
随后进入步骤220,从步骤210所确定的连通子图中选择一个子集。例如可以按照下列方式挑选该子集中的元素:首先选择总节点数在中等规模的连通子图。随后在这些中等规模的连通子图内统计转出金额和/或转出交易次数(以下又称为“出度”)或者转入金额和/或转入交易次数(以下又称为“入度”)较大的节点的数量,这些节点统称为可疑中心节点。最后将这些中等规模的连通子图内可疑中心节点数量较多的连通子图选入子集内。
在步骤220中,可以将出度(入度)大于阈值的节点视为可疑中心节点,该阈值的设定方式例如可以是:生成一个连通子图内的所有节点的出度(入度)的统计分布图,并且将统计分布图中的曲线转折点设定为出度(入度)的阈值。在步骤220中,还可以将可疑中心节点数量大于阈值的连通子图纳入子集。
随后进入步骤230,对子集内尚未进行社团划分操作的连通子图 执行社团划分操作。有关社团划分操作的详细描述将在下面给出。
接着进入步骤240,确定是否对于子集内的每个连通子图都实施了社团划分操作,如果是,则可以进入图1的步骤130,否则返回步骤230。
需要指出的是,在图2所示的方法中,步骤210和220是优选的步骤。也就是说,在一个实施方式中,可以直接对网络图执行如下所述的社团划分操作,或者对步骤210所确定的连通子图的每一个执行社团划分操作。
每一个连通子图都可以视为一个具有关联性质的交易群体。然而在这些众多的群体中,通常仅有一小部分涉及异常交易活动(例如洗钱)。而且一些非法交易活动的执行者还会刻意地将核心异常交易结构隐藏在大量的正常交易中,这进一步增加了异常交易的发现难度。本发明的发明人经过深入研究发现,如果对一个连通子图直接进行分析或社团划分操作,很可能出现的结果是虽然用于衡量该连通子图的异常交易的风险度量较低,然而实际上却隐藏着大量的异常交易。
针对上述情况,本发明的发明人创造性地引入下列方式来挖掘隐藏的异常交易:基于节点权重和交易时序对连通子图中的边的权重进行修正,然后利用为有向图专门定义的模块度,以迭代方式对边的权重修正后的连通子图进行社团划分,直到划分后该连通子图的模块度不再变化为止,由此完成该连通子图的社团划分。通过上述方式可以在连通子图内发现异常交易风险极大的社团或者多个异常交易风险较高的社团,从而大幅度提高异常交易的辨识度,并且还能够清楚地勾勒出核心的异常交易风险结构。
图3为可应用于图2所示实施例的社团划分算法的流程图,该算法基于上述方式。图3所示算法的操作对象为一个连通子图,但是这仅仅是示例性的,将整个网络图作为操作对象也是成立的。
图3所示的流程开始于步骤310。在该步骤中,利用节点权重对一个连通子图的每条边的权重进行修正或优化。优选地,可以利用一 个节点的交易金额、交易次数、出入度总数等交易信息来计算用于修正边的权重的节点权重。具体计算方式例如如下式(2)所示:
Figure PCTCN2018115141-appb-000004
这里,ω vj为节点j的节点权重,
Figure PCTCN2018115141-appb-000005
分别表示该节点j的总交易金额的标准化值、交易次数的标准化值以及出入度总数的标准化值,ω Mv、ω Cv、ω Dv为节点j的总交易金额、交易次数以及出入度总数的权重因子(例如每个权重因子可以都取值为1/3)。
对于第i条边而言,假设它的起始节点或金额转出节点为v i_in,目的节点或金额转入节点为v i_out,则利用第i条边的经节点权重修正后的权重W Ei变为:
Figure PCTCN2018115141-appb-000006
这里,ω Vi_in为初始节点的节点权重,ω Vi_out为目的节点的节点权重,W Bi为由式(1)确定的第i条边的初始权重。
对于一个连通子图内的每条边,都可以利用上式(2)和(3)来修正其权重,从而得到边的权重被利用节点权重修正过的连通子图。
随后进入步骤320。在该步骤中,对利用节点权重修正后的连通子图的边的权重进一步进行交易时序修正或优化。优选地,可以采用下列方式来作进一步的修正。
首先计算每个节点的平均转入和转出时间。例如对于连通子图内的任一节点A,假设有
Figure PCTCN2018115141-appb-000007
条边连入该节点,这
Figure PCTCN2018115141-appb-000008
条边中的第j条边连入该节点的时间为
Figure PCTCN2018115141-appb-000009
Figure PCTCN2018115141-appb-000010
条边中的第j条边连出该节点的时间为
Figure PCTCN2018115141-appb-000011
则节点A的平均连入时间为:
Figure PCTCN2018115141-appb-000012
节点A的平均连出时间为:
Figure PCTCN2018115141-appb-000013
随后确定与交易时序相关的权重修正系数。对于“先分散转入后集中转出”的情况(也即首先是多个节点向一个节点转账,接着由后者将汇集的金额集中转出的交易过程),从交易时序上考察,集中转出的那条边应该在多次分散转入的边之后形成。对于“先集中转入后分散转出”的情况(也即首先是一个节点接收一笔款项,然后由该节点将该笔款项向多个节点转账,最后多个节点将各自接收的款项转出的交易过程),从交易时序上考察,集中转入的那条边应该在多次分散转出的边之前形成。
在本实施例中,对于第i条边的两端的节点,根据交易的方向(即节点为交易的转出节点还是转入节点)定义不同的权重修正系数以用于基于交易时序的修正。具体而言,对于第i条边的初始节点src,其对应的权重修正系数θ 1按照下式确定:
Figure PCTCN2018115141-appb-000014
Figure PCTCN2018115141-appb-000015
Figure PCTCN2018115141-appb-000016
Figure PCTCN2018115141-appb-000017
这里,
Figure PCTCN2018115141-appb-000018
为初始节点src的入度,
Figure PCTCN2018115141-appb-000019
为初始节点src的出度,
Figure PCTCN2018115141-appb-000020
为初始节点src的平均连入时间,其可由式(4)确定,T src为初始节点src连出第j条边的时间,T R为规范化因子。
由上式(6)-(9)可见,对于满足条件
Figure PCTCN2018115141-appb-000021
Figure PCTCN2018115141-appb-000022
的边,其修正系数θ 1>1,其他情况下θ 1<1。
类似地,对于第i条边的目的节点dst,其对应的权重修正系数θ 2按照下式确定:
Figure PCTCN2018115141-appb-000023
Figure PCTCN2018115141-appb-000024
Figure PCTCN2018115141-appb-000025
Figure PCTCN2018115141-appb-000026
这里,
Figure PCTCN2018115141-appb-000027
为目的节点dst的出度,
Figure PCTCN2018115141-appb-000028
为目的节点dst的入度,
Figure PCTCN2018115141-appb-000029
为目的节点dst的平均连入时间,其可由式(5)确定,T dst为目的节点dst连入第j条边的时间,T R为规范化因子。
由上式(10)-(13)可见,对于满足条件
Figure PCTCN2018115141-appb-000030
Figure PCTCN2018115141-appb-000031
的边,其修正系数θ 2>1,其他情况下θ 2<1。
由此,对于第i条边,其权重可以按照下式进行基于交易时序的修正:
Figure PCTCN2018115141-appb-000032
这里,W Ei为步骤310中确定的第i条边的利用节点权重进行修正后的权重。
接着进入步骤330,在该步骤中,对经过步骤310和320的权重修正处理后的连通子图进行社团划分,从而将每个节点都划归到相应的社团内。
如上所述,在本实施例的网络图中,每条边为有向边。对于任意一条有向边i→j,令
Figure PCTCN2018115141-appb-000033
其中
Figure PCTCN2018115141-appb-000034
表示指向节点i的所有边的权重和,
Figure PCTCN2018115141-appb-000035
表示由节点i连出的所有边的权重和,k i表示节点j的所有边的权重和,k j表示节点j的所有边的权重和。
优选地,在本实施例中可以将模块度Q D定义为:
Figure PCTCN2018115141-appb-000036
Figure PCTCN2018115141-appb-000037
这里,如果节点i和节点j属于同一个社团,则δ(c i,c j)=1,否则δ(c i,c j)=0,A ij为有向网络的邻接权重矩阵中相应的值,如果存在边j→i,则A ij等于边的权重,否则为0,ΣW ec表示社团C内的边的权重 之和(包括社团内的点和社团外的点相连的边),m表示所有边的权重之和,Σ C代表对全部社团的求和,ΣM c表示仅对社团C内部矩阵M c的所有元素进行求和,M c具体表示如下:
Figure PCTCN2018115141-appb-000038
在本步骤中,优选地,可以采用与Louvain算法类似的迭代算法,利用上面定义的模块度来完成社团划分。
图4为可应用于图3所示实施例的迭代算法的流程图。
参见图4,在步骤410中,首先执行初始化处理,将一个连通子图中的每个节点划归到不同的社团中。
接着进入步骤420。在该步骤中,采用上式(15)定义的模块度,对于连通子图中的每个节点执行迭代操作。以该连通子图中的第i个节点为例,首先将节点i分配给它的每个邻居节点所属的社团,然后计算分配前与分配后的模块度变化值,从而得到与节点i相关联的一个或多个模块度变化值。在本实施例中,模块度变化值可以按照下式确定:
Figure PCTCN2018115141-appb-000039
Figure PCTCN2018115141-appb-000040
其中
Figure PCTCN2018115141-appb-000041
表示节点i与社团c内部节点的连边的权重之和。
在依照上式(18)和(19)得到到与节点i相关联的一个或多个模块度变化值之后,如果判断这些模块度变化值中的最大值maxΔQ D>0,则将节点i分配给与maxΔQ D对应的那个邻居节点所属的社团,否则使节点i保持在原社团不变。
接着进入步骤430。在该步骤中,确定所有节点归属社团的状态在本次执行步骤420前后是否发生变化,如果发生变化,则返回步骤420,否则进入步骤440。
在步骤440,按照下列方式对连通子图进行压缩:将属于同一社团的节点压缩为一个新节点,社团内节点之间的边的权重转化为新节点的环的权重,社团间的边权重转化为新节点间的边权重。
随后进入步骤450。在该步骤中,依照上式(15)-(17)确定步骤440中生成的压缩的连通子图的模块度,并且随后进入步骤460。
在步骤460,判断步骤450中确定的模块度与本次执行步骤440之前的连通子图的模块度之差是否小于预设的阈值,如果是,则进入步骤470,输出当前处理的连通子图的社团划分结果,否则返回步骤420。
图5为可应用于图1所示实施例的确定社团的风险量度的方法的流程图。为阐述方便起见,这里的描述以确定一个社团k的风险量度的过程为例。
图5所示的流程开始于步骤510。在该步骤中,确定时间段T m 期间待确定风险量度的社团的平均交易时间
Figure PCTCN2018115141-appb-000042
优选地,对于该社团在该段时间内的每笔交易,可以以最起始的一笔交易作为时间基准点来确定交易时间。
随后进入步骤520。对于该社团在该段时间内的每笔交易,确定其交易时间与平均交易时间之差的绝对值ΔT h,这里h为交易的索引号。
接着进入步骤530,根据ΔT h的取值将每笔交易归类到多个区间的相应区间中,并统计每个区间内的交易次数与该社团在时间段T m期间的总交易次数的比率。
随后进入步骤540,依照下式确定用于反映交易时间与异常交易之间相关性的交易时间熵H C
Figure PCTCN2018115141-appb-000043
这里n为区间的总数,P i表示第i个区间内的交易笔数与该社团在时间段T m期间的总交易笔数的比率。
由式(20)可见,在一个时间段内,如果一个社团内的交易时间熵越小,则表示交易活动的时间越集中,因此交易异常的可能性越大。
接着进入步骤550,确定该社团的整体风险因子。优选地,整体风险因子ψ k可以利用下式确定:
Figure PCTCN2018115141-appb-000044
这里
Figure PCTCN2018115141-appb-000045
为社团k内节点的数量的标准化值,
Figure PCTCN2018115141-appb-000046
为社团k在时间段T m期间的总交易次数的的标准化值,
Figure PCTCN2018115141-appb-000047
为社团k在时间段T m期间的总交易金额的的标准化值,
Figure PCTCN2018115141-appb-000048
为社团k内节点的平均度数的的标准化 值,
Figure PCTCN2018115141-appb-000049
为社团k在时间段T m期间的交易时间熵的标准化值,
Figure PCTCN2018115141-appb-000050
为权重值,可根据实际应用设定。
由式(21)计算得到的ψ k越大,则表明交易异常的风险度较大。
可选地但并非必须的,对于一个网络图或一个连通子图内的多个社团,可以按照图5所示方法确定的整体风险因子对它们进行从高到低的排序,其中前5%的社团被评级为I级可疑社团,介于5%~10%的社团被评级为II级可疑社团等。
在上面借助图1-5所述的实施例中,描述了用于识别一个时间段T m内的异常交易社团的方法。上述实施例也可以推广到多个时间段内异常交易社团的识别中。当需要对较长跨度的时间段内的交易活动进行监测时,考虑到社团可能的变化而将长跨度时间段分割为多个时间段来监测是有利的。
例如可以将一个较长跨度的时间段(例如一个星期、一个月或者半年等)分为n个时间段,然后在每个时间段内,分别采用上面借助图1-5所述的实施例来识别异常交易社团。考虑到数据量较大,优选地,可以采用下述增量式方法进行社团的划分。具体而言,在第一个时间段T i内完成社团划分后保留每个节点所对应的社团标签;随后,在对下一时间段T i+1进行社团划分时,取该时间段内的所有节点与上一时间段内的所有节点的交集,并且将交集部分的节点所对应的社团标签作为当前时间段的相关节点的初始标签,而将那些无社团标签的节点初始化为自身所属的社团,然后在此基础上执行社团划分操作。这种方式可以大大加快社团划分操作的收敛速度。
图6为按照本发明另一个实施例的用于识别异常交易社团的装置的框图。
图6所示的装置60包含存储器610、处理器620以及存储在存储器610上并可在处理器620上运行的计算机程序630,其中,计算机程序630通过在处理器620上运行以可执行如上借助图1-3所述实施例的方法。
图7为按照本发明另一个实施例的用于识别异常交易社团的装置的框图。
图7所示的装置70包含第一模块710、第二模块720和第三模块730,其中,第一模块710用于构建与多个账户相互间的交易事件相关的网络图,其中,所述网络图的每个节点代表所述多个账户的其中一个,并且以连接两个节点的边来指示在与这两个节点相关联的账户之间发生了交易,其中边的方向代表交易的方向;第二模块720用于从所述网络图确定为一个或多个社团;以及第三模块730用于根据社团的交易信息确定其相应的风险度量,该风险度量用于确定该社团是否属于异常交易社团。
按照本发明的一个方面,提供一种计算机可读存储介质,其上存储计算机程序,该程序被处理器执行时实现借助图1-3所述实施例的方法。
与现有技术相比,本发明的上述实施例具有下列优点:
1、不依赖已有案件信息,仅从海量交易中即能主动发现高风险的非法交易团伙。
2、通过创造性地将社团发现算法与动态洗钱模式相结合,形成了对于反洗钱具有特别针对性的时序有向社团发现算法,使得能够准确地进行洗钱意义上的社团划分。
3、能够对社团进行准确的异常交易风险量化评分,依照评分等级划分形成社团洗钱风险评级,业务人员能够根据该评级进行更加有目的性的反洗钱工作的开展。
4、通过动态分析多个时间跨度内的交易社团结构随时间的演化,能够确定高风险洗钱社团并分析其内在演化规律。
提供本文中提出的实施例和示例,以便最好地说明按照本技术及其特定应用的实施例,并且由此使本领域的技术人员能够实施和使用本发明。但是,本领域的技术人员将会知道,仅为了便于说明和举例而提供以上描述和示例。所提出的描述不是意在涵盖本发明的各个方 面或者将本发明局限于所公开的精确形式。
鉴于以上所述,本公开的范围通过以下权利要求书来确定。

Claims (11)

  1. 一种用于识别异常交易社团的方法,其特征在于,包含下列步骤:
    构建与多个账户相互间的交易事件相关的网络图,其中,所述网络图的每个节点代表所述多个账户的其中一个,并且以连接两个节点的边来指示在与这两个节点相关联的账户之间发生了交易,其中边的方向代表交易的方向;
    从所述网络图确定为一个或多个社团;以及
    根据社团的交易信息确定其相应的风险度量,该风险度量用于确定该社团是否属于异常交易社团。
  2. 如权利要求1所述的方法,其中,确定社团的步骤包括:
    从所述网络图确定一个或多个连通子图,其中,每个连通子图内的任意两个节点之间是连通的,并且两个连通子图之间无相连接的边;以及
    对连通子图执行社团划分操作。
  3. 如权利要求2所述的方法,其中,在执行社团划分的步骤中,对于任一连通子图,按照下列方式执行划分操作:
    基于节点权重和交易时序,对该连通子图中的边的权重进行修正;以及
    以迭代方式对该连通子图进行社团划分直到划分后该连通子图的模块度不再变化为止,由此完成该连通子图的社团划分。
  4. 如权利要求3所述的方法,其中,节点权重依赖于边两端的每个节点的交易金额、交易次数和出入度总数。
  5. 如权利要求3所述的方法,其中,所述交易时序依赖于边两端的每个节点的资金平均转入时间和资金平均转出时间。
  6. 如权利要求3所述的方法,其中,对于两个节点之间的边,其对模块度的贡献值与边的方向相关。
  7. 如权利要求1所述的方法,其中,所述交易信息包括每个社团内的每笔交易的时间、该社团的总交易数量和总交易金额。
  8. 如权利要求7所述的方法,其中,每个社团的风险度量包括该社团的交易时间熵和整体风险因子。
  9. 一种用于识别异常交易社团的装置,包含:
    第一模块,用于构建与多个账户相互间的交易事件相关的网络图,其中,所述网络图的每个节点代表所述多个账户的其中一个,并且以连接两个节点的边来指示在与这两个节点相关联的账户之间发生了交易,其中边的方向代表交易的方向;
    第二模块,用于从所述网络图确定为一个或多个社团;以及
    第三模块,用于根据社团的交易信息确定其相应的风险度量,该风险度量用于确定该社团是否属于异常交易社团。
  10. 一种用于识别异常交易社团的装置,包含存储器、处理器以及存储在所述存储器上并可在所述处理器上运行的计算机程序,其特征在于,执行如权利要求1-8中任意一项所述的方法。
  11. 一种计算机可读存储介质,其上存储计算机程序,其特征在于,该程序被处理器执行时实现如权利要求1-8中任意一项所述的方法。
PCT/CN2018/115141 2017-11-23 2018-11-13 用于识别异常交易社团的方法和装置 WO2019100967A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201711182339.8 2017-11-23
CN201711182339.8A CN108228706A (zh) 2017-11-23 2017-11-23 用于识别异常交易社团的方法和装置

Publications (1)

Publication Number Publication Date
WO2019100967A1 true WO2019100967A1 (zh) 2019-05-31

Family

ID=62652777

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/115141 WO2019100967A1 (zh) 2017-11-23 2018-11-13 用于识别异常交易社团的方法和装置

Country Status (3)

Country Link
CN (1) CN108228706A (zh)
TW (1) TWI759562B (zh)
WO (1) WO2019100967A1 (zh)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111401959A (zh) * 2020-03-18 2020-07-10 多点(深圳)数字科技有限公司 风险群体的预测方法、装置、计算机设备及存储介质
CN111612039A (zh) * 2020-04-24 2020-09-01 平安直通咨询有限公司上海分公司 异常用户识别的方法及装置、存储介质、电子设备
CN111612041A (zh) * 2020-04-24 2020-09-01 平安直通咨询有限公司上海分公司 异常用户识别方法及装置、存储介质、电子设备
CN111652718A (zh) * 2020-07-09 2020-09-11 平安银行股份有限公司 基于关系网络图的价值流向监控方法、装置、设备和介质
CN111740977A (zh) * 2020-06-16 2020-10-02 北京奇艺世纪科技有限公司 投票检测方法及装置、电子设备、计算机可读存储介质
CN112052404A (zh) * 2020-09-23 2020-12-08 西安交通大学 多源异构关系网络的群体发现方法、系统、设备及介质
CN112989272A (zh) * 2020-12-31 2021-06-18 中科院计算技术研究所大数据研究院 一种基于局部路径的社团发现算法
CN113393250A (zh) * 2021-06-09 2021-09-14 北京沃东天骏信息技术有限公司 一种信息处理方法及装置、存储介质
CN115048436A (zh) * 2022-06-01 2022-09-13 优米互动(北京)科技有限公司 基于可视图原理的高维金融时间序列的阶段划分方法

Families Citing this family (35)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108228706A (zh) * 2017-11-23 2018-06-29 中国银联股份有限公司 用于识别异常交易社团的方法和装置
CN109102151B (zh) * 2018-07-03 2021-08-31 创新先进技术有限公司 一种可疑群组识别方法和装置
CN109118053B (zh) * 2018-07-17 2022-04-05 创新先进技术有限公司 一种盗卡风险交易的识别方法和装置
CN109345252A (zh) * 2018-08-24 2019-02-15 阿里巴巴集团控股有限公司 一种线上交易控制方法、装置、及计算机设备
CN109146669A (zh) * 2018-08-24 2019-01-04 阿里巴巴集团控股有限公司 异常资金调度计划的检测方法、装置及服务器
CN109947865B (zh) * 2018-09-05 2023-06-30 中国银联股份有限公司 商户分类方法及商户分类系统
CN109272323B (zh) * 2018-09-14 2022-03-04 创新先进技术有限公司 一种风险交易识别方法、装置、设备及介质
CN111046237B (zh) * 2018-10-10 2024-04-05 京东科技控股股份有限公司 用户行为数据处理方法、装置、电子设备及可读介质
CN109460664B (zh) * 2018-10-23 2022-05-03 北京三快在线科技有限公司 风险分析方法、装置、电子设备及计算机可读介质
CN109598511B (zh) * 2018-11-05 2023-06-20 创新先进技术有限公司 一种账户风险识别方法、装置及设备
CN109615521A (zh) * 2018-12-26 2019-04-12 天翼电子商务有限公司 基于营销反套利模型的反套利识别方法、系统及服务器
CN109872232A (zh) * 2019-01-04 2019-06-11 平安科技(深圳)有限公司 涉及非法所得合法化行为的账户分类方法、装置、计算机设备及存储介质
CN111754342B (zh) * 2019-03-26 2024-05-24 众安信息技术服务有限公司 一种获得区块链加密货币流通速度的方法、系统及装置
CN111951021A (zh) * 2019-05-15 2020-11-17 财付通支付科技有限公司 一种可疑社团的发现方法和装置、存储介质及计算机设备
CN110222297B (zh) * 2019-06-19 2021-07-23 武汉斗鱼网络科技有限公司 一种标签用户的识别方法以及相关设备
CN110490730B (zh) * 2019-08-21 2022-07-26 北京顶象技术有限公司 异常资金聚集行为检测方法、装置、设备及存储介质
CN110544104B (zh) * 2019-09-04 2024-01-23 北京趣拿软件科技有限公司 帐号的确定方法、装置、存储介质及电子装置
CN110705995B (zh) * 2019-10-10 2022-08-30 支付宝(杭州)信息技术有限公司 数据标签化方法和装置
CN110717758B (zh) * 2019-10-10 2021-04-13 支付宝(杭州)信息技术有限公司 异常交易识别方法和装置
CN111177477B (zh) * 2019-12-06 2023-06-20 东软集团股份有限公司 一种可疑群组的确定方法、装置及设备
CN111161063A (zh) * 2019-12-12 2020-05-15 厦门市美亚柏科信息股份有限公司 基于图计算的资金账号识别方法及计算机可读存储介质
CN112990919A (zh) * 2019-12-17 2021-06-18 中国银联股份有限公司 一种信息处理的方法及装置
CN111242763A (zh) * 2020-01-07 2020-06-05 北京明略软件系统有限公司 一种目标用户群的确定方法及装置
CN111340622A (zh) * 2020-02-21 2020-06-26 中国银联股份有限公司 一种异常交易集群的检测方法及装置
CN113313505B (zh) * 2020-02-25 2023-07-25 中国移动通信集团浙江有限公司 异常定位方法、装置及计算设备
CN111476662A (zh) * 2020-04-13 2020-07-31 中国工商银行股份有限公司 反洗钱识别方法及装置
CN111770047B (zh) * 2020-05-07 2022-09-23 拉扎斯网络科技(上海)有限公司 异常群体的检测方法、装置及设备
CN111339376B (zh) * 2020-05-15 2020-10-27 支付宝(杭州)信息技术有限公司 用于网络节点聚类的方法以及装置
CN111831923A (zh) * 2020-07-14 2020-10-27 北京芯盾时代科技有限公司 识别关联的特定账户的方法、装置及存储介质
CN112381544B (zh) * 2020-11-16 2022-09-02 支付宝(杭州)信息技术有限公司 子图确定方法、装置和电子设备
CN112491900B (zh) * 2020-11-30 2023-04-18 中国银联股份有限公司 异常节点识别方法、装置、设备及介质
CN113487427A (zh) * 2021-04-20 2021-10-08 微梦创科网络科技(中国)有限公司 一种交易风险识别方法、装置及系统
CN113554308B (zh) * 2021-07-23 2024-05-28 中信银行股份有限公司 用户社团的划分与风险用户的识别方法、装置及电子设备
CN113837874B (zh) * 2021-11-22 2022-04-12 北京芯盾时代科技有限公司 一种数据的识别方法、装置、存储介质及电子设备
CN116340090A (zh) * 2023-02-09 2023-06-27 中科南京软件技术研究院 基于交互序列的软件识别方法、装置、设备及存储介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105335855A (zh) * 2014-08-06 2016-02-17 阿里巴巴集团控股有限公司 一种交易风险识别方法及装置
CN105931046A (zh) * 2015-12-16 2016-09-07 中国银联股份有限公司 一种可疑交易节点集合侦测方法及装置
CN105988998A (zh) * 2015-01-27 2016-10-05 阿里巴巴集团控股有限公司 关系网络构建方法及装置
CN108228706A (zh) * 2017-11-23 2018-06-29 中国银联股份有限公司 用于识别异常交易社团的方法和装置

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TW201123055A (en) * 2009-12-31 2011-07-01 Yao-Lang Guo Method of monitoring and evaluating securities trade safety.
CN104751566B (zh) * 2013-12-30 2018-11-27 中国银联股份有限公司 监测伪卡风险的方法和实现该方法的交易处理系统
CN107103171B (zh) * 2016-02-19 2020-09-25 阿里巴巴集团控股有限公司 机器学习模型的建模方法及装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105335855A (zh) * 2014-08-06 2016-02-17 阿里巴巴集团控股有限公司 一种交易风险识别方法及装置
CN105988998A (zh) * 2015-01-27 2016-10-05 阿里巴巴集团控股有限公司 关系网络构建方法及装置
CN105931046A (zh) * 2015-12-16 2016-09-07 中国银联股份有限公司 一种可疑交易节点集合侦测方法及装置
CN108228706A (zh) * 2017-11-23 2018-06-29 中国银联股份有限公司 用于识别异常交易社团的方法和装置

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111401959B (zh) * 2020-03-18 2023-09-29 多点(深圳)数字科技有限公司 风险群体的预测方法、装置、计算机设备及存储介质
CN111401959A (zh) * 2020-03-18 2020-07-10 多点(深圳)数字科技有限公司 风险群体的预测方法、装置、计算机设备及存储介质
CN111612039A (zh) * 2020-04-24 2020-09-01 平安直通咨询有限公司上海分公司 异常用户识别的方法及装置、存储介质、电子设备
CN111612041A (zh) * 2020-04-24 2020-09-01 平安直通咨询有限公司上海分公司 异常用户识别方法及装置、存储介质、电子设备
CN111612041B (zh) * 2020-04-24 2023-10-13 平安直通咨询有限公司上海分公司 异常用户识别方法及装置、存储介质、电子设备
CN111612039B (zh) * 2020-04-24 2023-09-29 平安直通咨询有限公司上海分公司 异常用户识别的方法及装置、存储介质、电子设备
CN111740977A (zh) * 2020-06-16 2020-10-02 北京奇艺世纪科技有限公司 投票检测方法及装置、电子设备、计算机可读存储介质
CN111740977B (zh) * 2020-06-16 2022-06-21 北京奇艺世纪科技有限公司 投票检测方法及装置、电子设备、计算机可读存储介质
CN111652718A (zh) * 2020-07-09 2020-09-11 平安银行股份有限公司 基于关系网络图的价值流向监控方法、装置、设备和介质
CN112052404A (zh) * 2020-09-23 2020-12-08 西安交通大学 多源异构关系网络的群体发现方法、系统、设备及介质
CN112052404B (zh) * 2020-09-23 2023-08-15 西安交通大学 多源异构关系网络的群体发现方法、系统、设备及介质
CN112989272A (zh) * 2020-12-31 2021-06-18 中科院计算技术研究所大数据研究院 一种基于局部路径的社团发现算法
CN112989272B (zh) * 2020-12-31 2024-02-27 中科院计算技术研究所大数据研究院 一种基于局部路径的社团发现算法
CN113393250A (zh) * 2021-06-09 2021-09-14 北京沃东天骏信息技术有限公司 一种信息处理方法及装置、存储介质
CN115048436A (zh) * 2022-06-01 2022-09-13 优米互动(北京)科技有限公司 基于可视图原理的高维金融时间序列的阶段划分方法

Also Published As

Publication number Publication date
TW201926204A (zh) 2019-07-01
CN108228706A (zh) 2018-06-29
TWI759562B (zh) 2022-04-01

Similar Documents

Publication Publication Date Title
WO2019100967A1 (zh) 用于识别异常交易社团的方法和装置
US11263566B2 (en) Seasonality validation and determination of patterns
US9043647B2 (en) Fault detection and localization in data centers
CN110490730B (zh) 异常资金聚集行为检测方法、装置、设备及存储介质
WO2018014610A1 (zh) 基于c4.5决策树算法的特定用户挖掘系统及其方法
CN109873812A (zh) 异常检测方法、装置及计算机设备
Wang et al. Detecting potential collusive cliques in futures markets based on trading behaviors from real data
WO2020100108A1 (en) Systems and method for scoring entities and networks in a knowledge graph
CN102801629B (zh) 一种流量矩阵的估计方法
US11568179B2 (en) Selecting an algorithm for analyzing a data set based on the distribution of the data set
WO2019200739A1 (zh) 数据欺诈识别方法、装置、计算机设备和存储介质
JP6553816B2 (ja) ユーザデータ共有方法及び装置
Zhou et al. Asymmetric fuzzy preference relations based on the generalized sigmoid scale and their application in decision making involving risk appetites
CN111145027A (zh) 一种疑似洗钱交易识别方法及装置
CN114187112A (zh) 账户风险模型的训练方法和风险用户群体的确定方法
CN109977131A (zh) 一种房型匹配系统
CN113112186A (zh) 一种企业评估方法、装置及设备
CN107423319B (zh) 一种垃圾网页检测方法
CN109509019A (zh) 房地产项目经营状况监控应用方法、系统及云应用系统
Yan et al. The impact of capital market opening on stock market stability: Based on D-MST method complex network perspective
TW202133089A (zh) 最佳化驅動決策之方法及其電腦程式產品
TWM568442U (zh) Golden flow grouping system
CN109919811A (zh) 基于大数据的保险代理人培养方案生成方法及相关设备
Feng et al. Systemic risk models for disjoint and overlapping groups with equilibrium strategies
JP2017509047A (ja) 特許資産を用いるテクノロジーセクター規模の報告

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18880965

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18880965

Country of ref document: EP

Kind code of ref document: A1