CN113641906A - System, method, device, processor and medium for realizing similar target person identification processing based on fund transaction relation data - Google Patents
System, method, device, processor and medium for realizing similar target person identification processing based on fund transaction relation data Download PDFInfo
- Publication number
- CN113641906A CN113641906A CN202110940032.XA CN202110940032A CN113641906A CN 113641906 A CN113641906 A CN 113641906A CN 202110940032 A CN202110940032 A CN 202110940032A CN 113641906 A CN113641906 A CN 113641906A
- Authority
- CN
- China
- Prior art keywords
- node
- similar
- fund
- fund transaction
- transaction
- 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
- 238000000034 method Methods 0.000 title claims abstract description 64
- 238000012545 processing Methods 0.000 title claims abstract description 61
- 239000013598 vector Substances 0.000 claims abstract description 50
- 238000007781 pre-processing Methods 0.000 claims abstract description 11
- 230000002457 bidirectional effect Effects 0.000 claims abstract description 10
- 230000008569 process Effects 0.000 claims description 17
- 238000012216 screening Methods 0.000 claims description 13
- 238000004422 calculation algorithm Methods 0.000 claims description 11
- 238000012549 training Methods 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 6
- 230000001172 regenerating effect Effects 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 3
- 238000011160 research Methods 0.000 abstract description 6
- 230000006399 behavior Effects 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008520 organization Effects 0.000 description 2
- 230000007704 transition Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Evolutionary Computation (AREA)
- Finance (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- General Business, Economics & Management (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to a system for realizing similar target person identification processing based on fund transaction relation data, wherein the system comprises a data acquisition and processing module, a data preprocessing module and a data processing module, wherein the data acquisition and processing module is used for constructing a fund transaction relation data set by collecting fund transaction relation data and carrying out data preprocessing; the fund transaction network graph model generating module is used for constructing a fund transaction network graph model to generate a node walking sequence; the node embedded vector generation module is used for generating a bidirectional embedded vector of each node; and the similar target determination generation module is used for determining the target object of the node which is most similar to the known node. The invention also relates to a corresponding method, device, processor and storage medium thereof. By adopting the system, the method, the device, the processor and the storage medium thereof, the similar target object can be quickly detected, so that similar unknown target personnel can be quickly detected under the condition of a known part of target personnel nodes, and a more efficient research and judgment method support is provided for relevant departments.
Description
Technical Field
The invention relates to the technical field of deep learning, in particular to the technical field of data studying and judging analysis, and specifically relates to a system, a method, a device, a processor and a computer readable storage medium for realizing similar target person identification processing based on fund transaction relation data.
Background
With the rapid development of computer and internet technologies in recent years, new and intelligent economic illegal events are in an increasing trend, the organization trend of related target personnel main bodies is obvious, and the quantity and dimension of related events related to the events are increased explosively. The related data is usually extremely large in size, reasonably mixed with abnormal transactions and difficult to process by the existing means; the incidence relation among the data is various, and the overall data presents the non-structural characteristic. The traditional detection technology mainly depends on manual means to search abnormal records and track related clues according to experience, needs a great deal of time and labor, and cannot ensure the efficiency and quality of judgment due to various subjective factors. Such an approach appears quite weak when faced with large-scale data.
Under such circumstances, it is desirable to introduce the existing artificial intelligence technology into the detection of relevant events, to draw pictures of target persons and their relevant behaviors through man-machine integration, to allow a machine to complete time-consuming and labor-consuming data processing, and to feed back the initial processing results to a technician for further analysis. Compared with the traditional method, the data can be rapidly processed and analyzed through man-machine combination, and the abnormal records in the data can be objectively found and the target clues can be tracked from the whole data.
The target person and the related behaviors thereof are accurately represented, and a related algorithm based on graph learning or network embedding can be adopted. The map is a general data structure with strong expression capacity, and can be used for describing a plurality of problems in actual investigation, such as a fund flow network among target personnel, network communication in a telecommunication scene, a traffic flow network in transportation and the like. The graph learning can capture key information and hidden features in graph data, and has accurate and efficient analysis and processing capacity when the graph learning is oriented to a large-scale and multi-dimensional heterogeneous network. The current popular graph learning algorithms such as community discovery, graph embedding, network visualization and the like can visually and effectively assist technicians to quickly lock related target personnel and master criminal organization structures, and the aims of quick early warning, accurate striking and effective containment are achieved.
Graphs (or networks) composed of real data tend to be high-dimensional, unstructured. The high-dimensional representation is characterized in two aspects, firstly, when the scale of the graph is large, a high-dimensional space is needed to describe or characterize the structure of the graph; secondly, when the amount of information contained in the elements in the graph is large (such as a heterogeneous network), a more high-dimensional mathematical model needs to be constructed. Graph embedding is a hot spot direction of graph learning in recent years, and can embed high-dimensional information of a graph into a low-dimensional vector space, retain important information such as original structure, topology and attributes, and convert unstructured graph data into a structured low-dimensional vector.
In a general case study and judgment analysis scene, the analysis of a fund network plays a very important role, and after a case handling personnel preliminarily determines a few target personnel, the case handling personnel legally collects fund transaction data related to the target personnel. In recent years, with the development of internet technology, various mobile payment and online transactions become very common, which directly results in the rapid increase of the scale of bad fund transaction data, and also brings new challenges to technicians analyzing the data, namely how to rapidly develop unknown target personnel with similar behaviors with the mastered target personnel on the basis of independent of manual experience.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a system, a method, a device, a processor and a computer readable storage medium thereof for realizing similar target person identification processing based on fund transaction relation data, which have high efficiency, low time consumption and low experience requirement on research and judgment persons.
To achieve the above objects, the system, method, apparatus, processor and computer readable storage medium thereof of the present invention for implementing similar target person identification processing based on fund transaction relationship data are as follows:
the system for realizing similar target person identification processing based on fund transaction relation data is mainly characterized by comprising the following steps:
the data acquisition and processing module is used for constructing a fund transaction relation data set by collecting fund transaction relation data and carrying out data preprocessing on the fund transaction relation data set;
the fund transaction network graph model generating module is connected with the data acquisition and processing module and used for constructing a fund transaction network graph model for the fund transaction relation data set subjected to data preprocessing and generating a node walking sequence according to the transaction direction;
the node embedded vector generating module is connected with the fund transaction network graph generating module and used for generating a bidirectional embedded vector of each node;
and the similar target determining and generating module is connected with the node embedded vector generating module and used for determining the target object of the similar node closest to the known node by calculating the distance between the bidirectional embedded vectors of all the nodes.
Preferably, the fund transaction network graph model generation module specifically includes:
the fund transaction network graph generating and processing unit is used for generating a fund transaction network graph model according to the fund transaction relation data set and generating a new transaction sub-network through the comparison relation between each node and a preset threshold value node; and
and the node walking sequence generating unit is connected with the fund transaction network graph generating and processing unit and is used for generating a corresponding node walking sequence for the nodes in the transaction sub-network according to a node2vec algorithm.
Preferably, the fund transaction network graph generation processing unit performs the following operations:
and calculating directed edge weights according to the transaction amount among all the nodes by using the fund transaction network graph model, screening out the determined nodes with fund inflow weights and fund outflow weights larger than preset thresholds, and regenerating the transaction sub-networks among the determined nodes.
Preferably, the node-embedded vector generating module specifically includes:
the node pair set generating unit is connected with the node walking sequence generating unit and used for generating a corresponding node pair set by the node walking sequence according to the fund flow direction; and
and the embedded vector generating unit is connected with the node pair set generating unit and used for carrying out network training on each node pair in the node pair set by using a Word2Vec method aiming at node pairs in different directions and extracting the embedded vector of each node pair in two directions.
Preferably, the similar object determining and generating module specifically includes:
the similar node calculation processing unit is used for determining similar nodes by calculating embedded vector distances between the known nodes and all unknown nodes; and
and the target object determination processing unit is connected with the similar node calculation processing unit and used for determining a similar node set which is closest to the target node according to the sequencing and screening of the similar nodes according to the embedded vector distance.
The method for realizing similar target person identification processing based on fund transaction relation data by using the system is mainly characterized by comprising the following steps of:
(1) the data acquisition and processing module is used for collecting fund transaction relationship data to construct a fund transaction relationship data set, and data preprocessing is carried out on the fund transaction relationship data set;
(2) constructing a fund transaction network graph model according to the fund transaction relation data set, and generating a node walking sequence according to the transaction direction;
(3) generating corresponding bidirectional embedded vectors for each node in the node walking sequence by using a Word2Vec method;
(4) and determining the target object of the similar node closest to the known node according to the distance between the embedded vectors of the known node and the target node.
Preferably, the step (2) specifically comprises the following steps:
(2.1) constructing a fund transaction network graph model according to the fund transaction relation data set;
(2.2) calculating the directed edge weight between every two nodes by using the transaction amount in the fund transaction relation data set;
(2.3) screening out the determined nodes with the fund inflow weight and the fund outflow weight larger than the preset threshold weight of the system;
(2.4) reconstructing said determined node into a trading sub-network;
(2.5) generating a corresponding node walk sequence for each node in the trading sub-network based on a node2vec algorithm.
Preferably, the step (3) specifically includes the following steps:
(3.1) dividing the node walking sequence into a node pair set according to the fund flow direction;
and (3.2) carrying out network training on each node pair in the node pair set by using a Word2Vec method aiming at node pairs in different directions, and extracting embedded vectors of each node pair in two directions.
Preferably, the step (4) specifically includes the following steps:
(4.1) calculating embedded vector distances between the known nodes and all unknown nodes, and determining similar nodes;
(4.2) sequencing and screening the similar nodes according to the distance of the embedded vectors, and determining a similar node set which is closest to a target node;
and (4.3) determining a target object according to the similar node set.
The device for realizing similar target person identification processing based on fund transaction relation data is mainly characterized by comprising the following steps:
a processor configured to execute computer-executable instructions;
a memory storing one or more computer-executable instructions that, when executed by the processor, perform the steps of the method of similar target person identification processing based on funds transaction relationship data described above.
The processor for similar target person identification processing based on fund transaction relationship data is mainly characterized in that the processor is configured to execute computer executable instructions, and when the computer executable instructions are executed by the processor, the steps of the method for similar target person identification processing based on fund transaction relationship data are realized.
The computer-readable storage medium is primarily characterized by having stored thereon a computer program executable by a processor to perform the steps of the above-described method for similar target person identification processing based on funds transaction relationship data.
The system, the method, the device, the processor and the computer readable storage medium for realizing similar target person identification processing based on the fund transaction relation data provide an unknown target person with high efficiency, low time consumption and low experience requirement on research and judgment personnel based on a specific algorithm model, can quickly and accurately discover similar behaviors to the mastered target person from large-scale fund transaction data, provide more efficient research and judgment method support for related departments, and have wider application range and outstanding practicability.
Drawings
FIG. 1 is a block diagram of the system for implementing similar target person identification processing based on fund transaction relationship data according to the present invention.
FIG. 2 is a flow chart of the method of the present invention for implementing a similar target person identification process based on funding transaction relationship data.
Detailed Description
In order to more clearly describe the technical contents of the present invention, the following further description is given in conjunction with specific embodiments.
Before describing in detail embodiments that are in accordance with the present invention, it should be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Referring to fig. 1, the system for implementing similar target person identification processing based on fund transaction relationship data includes:
the data acquisition and processing module is used for constructing a fund transaction relation data set by collecting fund transaction relation data and carrying out data preprocessing on the fund transaction relation data set;
the fund transaction network graph model generating module is connected with the data acquisition and processing module and used for constructing a fund transaction network graph model for the fund transaction relation data set subjected to data preprocessing and generating a node walking sequence according to the transaction direction;
the node embedded vector generating module is connected with the fund transaction network graph generating module and used for generating a bidirectional embedded vector of each node;
and the similar target determining and generating module is connected with the node embedded vector generating module and used for determining the target object of the similar node closest to the known node by calculating the distance between the bidirectional embedded vectors of all the nodes.
As a preferred embodiment of the present invention, the fund transaction network graph model generation module specifically includes:
the fund transaction network graph generating and processing unit is used for generating a fund transaction network graph model according to the fund transaction relation data set and generating a new transaction sub-network through the comparison relation between each node and a preset threshold value node; and
and the node walking sequence generating unit is connected with the fund transaction network graph generating and processing unit and is used for generating a corresponding node walking sequence for the nodes in the transaction sub-network according to a node2vec algorithm.
As a preferred embodiment of the present invention, the fund transaction network graph generation processing unit specifically performs the following operations:
and calculating directed edge weights according to the transaction amount among all the nodes by using the fund transaction network graph model, screening out the determined nodes with fund inflow weights and fund outflow weights larger than preset thresholds, and regenerating the transaction sub-networks among the determined nodes.
As a preferred embodiment of the present invention, the node-embedded vector generation module specifically includes:
the node pair set generating unit is connected with the node walking sequence generating unit and used for generating a corresponding node pair set by the node walking sequence according to the fund flow direction; and
and the embedded vector generating unit is connected with the node pair set generating unit and used for carrying out network training on each node pair in the node pair set by using a Word2Vec method aiming at node pairs in different directions and extracting the embedded vector of each node pair in two directions.
As a preferred embodiment of the present invention, the similar object determination and generation module specifically includes:
the similar node calculation processing unit is used for determining similar nodes by calculating embedded vector distances between the known nodes and all unknown nodes; and
and the target object determination processing unit is connected with the similar node calculation processing unit and used for determining a similar node set which is closest to the target node according to the sequencing and screening of the similar nodes according to the embedded vector distance.
The method for realizing similar target person identification processing based on fund transaction relation data by utilizing the system comprises the following steps:
(1) the data acquisition and processing module is used for collecting fund transaction relationship data to construct a fund transaction relationship data set, and data preprocessing is carried out on the fund transaction relationship data set;
(2) constructing a fund transaction network graph model according to the fund transaction relation data set, and generating a node walking sequence according to the transaction direction;
(3) generating corresponding bidirectional embedded vectors for each node in the node walking sequence by using a Word2Vec method;
(4) and determining the target object of the similar node closest to the known node according to the distance between the embedded vectors of the known node and the target node.
As a preferred embodiment of the present invention, the step (2) specifically comprises the following steps:
(2.1) constructing a fund transaction network graph model according to the fund transaction relation data set;
(2.2) calculating the directed edge weight between every two nodes by using the transaction amount in the fund transaction relation data set;
(2.3) screening out the determined nodes with the fund inflow weight and the fund outflow weight larger than the preset threshold weight of the system;
(2.4) reconstructing said determined node into a trading sub-network;
(2.5) generating a corresponding node walk sequence for each node in the trading sub-network based on a node2vec algorithm.
As a preferred embodiment of the present invention, the step (3) specifically comprises the following steps:
(3.1) dividing the node walking sequence into a node pair set according to the fund flow direction;
and (3.2) carrying out network training on each node pair in the node pair set by using a Word2Vec method aiming at node pairs in different directions, and extracting embedded vectors of each node pair in two directions.
As a preferred embodiment of the present invention, the step (4) specifically comprises the following steps:
(4.1) calculating embedded vector distances between the known nodes and all unknown nodes, and determining similar nodes;
(4.2) sequencing and screening the similar nodes according to the distance of the embedded vectors, and determining a similar node set which is closest to a target node;
and (4.3) determining a target object according to the similar node set.
The apparatus for implementing a similar target person identification process based on funding transaction relationship data, wherein the apparatus comprises:
a processor configured to execute computer-executable instructions;
a memory storing one or more computer-executable instructions that, when executed by the processor, perform the steps of the method of similar target person identification processing based on funds transaction relationship data described above.
The processor for similar target person identification processing based on fund transaction relationship data is configured to execute computer executable instructions, and when the computer executable instructions are executed by the processor, the steps of the method for similar target person identification processing based on fund transaction relationship data are realized.
The computer readable storage medium has stored thereon a computer program executable by a processor to perform the steps of the method for similar target person identification processing based on funds transaction relationship data as described above.
In a specific embodiment of the present invention, the method for implementing similar target person identification processing based on fund transaction relationship data specifically includes:
(1) after the fund transaction data is cleaned, a directed fund transaction network graph is constructed according to the fund transaction data, each node in the network represents one person, directed edges between the nodes represent that fund transaction records exist between the two persons, the direction of the edges represents the direction of fund flow, and the weight of the edges is defined according to the total fund transaction amount. And screening out nodes with the fund inflow weight or the fund outflow weight larger than a certain threshold value, and regenerating a trading sub-network between the nodes. A series of node walk sequences are generated based on the node2vec algorithm according to the generated trading sub-network.
Assuming that the current node is v and the previous node is t, that is, the node sequence [ v, t ] is generated, the transition probability of selecting the next node x from the neighboring nodes of the current node v is:
if t is equal to x, then the probability of sampling x is
If t is connected to x, then the probability of sampling x is 1, and if t is not connected to x, then the probability of sampling x is
And finding all neighbor nodes of the current node from the addition of the initial node u, selecting a neighbor s according to the transition probability sampling, adding the neighbor to walk, generating a walk sequence walk for each node in the graph by the method, and adding the walk to the walk, wherein the walk is used for storing the walk sequence.
(2) Aiming at the generated walk sequence, dividing a target node and a source node in two directions according to the fund direction, and assuming that the current walk is [ x ]1,x2,…xi,xi+1,…,xl]The current node is xiThe window length is w;
sample pairs may be generated from the fund flow:
sample pairs may be generated from the fund inflow
Sample pairs based on fund flow
Generating training samples, where xiAnd (3) as input nodes, and other nodes as target nodes, entering a Word2Vec deep neural network, and training to generate a fund source characterization embedding vector of each node.
Sample pairs based on fund inflow
Generating training samples, where xiAnd as input nodes, other nodes are used as target nodes, a Word2Vec deep neural network is used, and the fund direction of each node is generated by training to represent the embedded vector.
(3) Computing embedded vectors Ex for all nodes in a networkii-inner product distance between 1,2, … n and the known node embedding vector Ex:
d(Ex,Exi)＝(Ex·Exi)
the sorting selects m nodes [ x ] with minimum distance1,x12,…xm]And the detected target person node set has behavior similarity with the known node x.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and alternate implementations are included within the scope of the preferred embodiment of the present invention in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present invention.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by suitable instruction execution devices.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, and the program may be stored in a computer readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present invention may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a separate product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc.
In the description herein, references to the description of terms "an embodiment," "some embodiments," "an example," "a specific example," or "an embodiment," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
The system, the method, the device, the processor and the computer readable storage medium for realizing similar target person identification processing based on the fund transaction relation data provide an unknown target person with high efficiency, low time consumption and low experience requirement on research and judgment personnel based on a specific algorithm model, can quickly and accurately discover similar behaviors to the mastered target person from large-scale fund transaction data, provide more efficient research and judgment method support for related departments, and have wider application range and outstanding practicability.
In this specification, the invention has been described with reference to specific embodiments thereof. It will, however, be evident that various modifications and changes may be made thereto without departing from the broader spirit and scope of the invention. The specification and drawings are, accordingly, to be regarded in an illustrative rather than a restrictive sense.
Claims (12)
1. A system for implementing a similar target person identification process based on funding transaction relationship data, the system comprising:
the data acquisition and processing module is used for constructing a fund transaction relation data set by collecting fund transaction relation data and carrying out data preprocessing on the fund transaction relation data set;
the fund transaction network graph model generating module is connected with the data acquisition and processing module and used for constructing a fund transaction network graph model for the fund transaction relation data set subjected to data preprocessing and generating a node walking sequence according to the transaction direction;
the node embedded vector generating module is connected with the fund transaction network graph generating module and used for generating a bidirectional embedded vector of each node;
and the similar target determining and generating module is connected with the node embedded vector generating module and used for determining the target object of the similar node closest to the known node by calculating the distance between the bidirectional embedded vectors of all the nodes.
2. The system for achieving similar target person identification processing based on fund transaction relationship data as claimed in claim 1, wherein the fund transaction network map model generating module comprises:
the fund transaction network graph generating and processing unit is used for generating a fund transaction network graph model according to the fund transaction relation data set and generating a new transaction sub-network through the comparison relation between each node and a preset threshold value node; and
and the node walking sequence generating unit is connected with the fund transaction network graph generating and processing unit and is used for generating a corresponding node walking sequence for the nodes in the transaction sub-network according to a node2vec algorithm.
3. The system for achieving similar target person identification processing based on fund transaction relationship data as claimed in claim 2, wherein the fund transaction network map generation processing unit performs the following operations:
and calculating directed edge weights according to the transaction amount among all the nodes by using the fund transaction network graph model, screening out the determined nodes with fund inflow weights and fund outflow weights larger than preset thresholds, and regenerating the transaction sub-networks among the determined nodes.
4. The system for achieving similar target person identification processing based on fund transaction relationship data as claimed in claim 3, wherein the node embedding vector generation module specifically comprises:
the node pair set generating unit is connected with the node walking sequence generating unit and used for generating a corresponding node pair set by the node walking sequence according to the fund flow direction; and
and the embedded vector generating unit is connected with the node pair set generating unit and used for carrying out network training on each node pair in the node pair set by using a Word2Vec method aiming at node pairs in different directions and extracting the embedded vector of each node pair in two directions.
5. The system for achieving similar objective person identification processing based on fund transaction relationship data as claimed in claim 4, wherein the similar objective determination generating module comprises:
the similar node calculation processing unit is used for determining similar nodes by calculating embedded vector distances between the known nodes and all unknown nodes; and
and the target object determination processing unit is connected with the similar node calculation processing unit and used for determining a similar node set which is closest to the target node according to the sequencing and screening of the similar nodes according to the embedded vector distance.
6. A method for implementing a similar target person identification process based on funding transaction relationship data using the system of claim 1, the method comprising the steps of:
(1) the data acquisition and processing module is used for collecting fund transaction relationship data to construct a fund transaction relationship data set, and data preprocessing is carried out on the fund transaction relationship data set;
(2) constructing a fund transaction network graph model according to the fund transaction relation data set, and generating a node walking sequence according to the transaction direction;
(3) generating corresponding bidirectional embedded vectors for each node in the node walking sequence by using a Word2Vec method;
(4) and determining the target object of the similar node closest to the known node according to the distance between the embedded vectors of the known node and the target node.
7. The method for implementing a similar target person identification process based on fund transaction relationship data as claimed in claim 6, wherein said step (2) comprises the following steps:
(2.1) constructing a fund transaction network graph model according to the fund transaction relation data set;
(2.2) calculating the directed edge weight between every two nodes by using the transaction amount in the fund transaction relation data set;
(2.3) screening out the determined nodes with the fund inflow weight and the fund outflow weight larger than the preset threshold weight of the system;
(2.4) reconstructing said determined node into a trading sub-network;
(2.5) generating a corresponding node walk sequence for each node in the trading sub-network based on a node2vec algorithm.
8. The method for implementing a similar target person identification process based on fund transaction relationship data as claimed in claim 7, wherein said step (3) comprises the following steps:
(3.1) dividing the node walking sequence into a node pair set according to the fund flow direction;
and (3.2) carrying out network training on each node pair in the node pair set by using a Word2Vec method aiming at node pairs in different directions, and extracting embedded vectors of each node pair in two directions.
9. The method for implementing a similar target person identification process based on fund transaction relationship data as claimed in claim 8, wherein said step (4) comprises the following steps:
(4.1) calculating embedded vector distances between the known nodes and all unknown nodes, and determining similar nodes;
(4.2) sequencing and screening the similar nodes according to the distance of the embedded vectors, and determining a similar node set which is closest to a target node;
and (4.3) determining a target object according to the similar node set.
10. An apparatus for implementing a similar target person identification process based on funding transaction relationship data, the apparatus comprising:
a processor configured to execute computer-executable instructions;
a memory storing one or more computer-executable instructions that, when executed by the processor, perform the steps of the method of any one of claims 6 to 9 for a similar target person identification process based on funds transaction relationship data.
11. A processor for a similar target person identification process based on funds transaction relationship data, wherein the processor is configured to execute computer executable instructions which, when executed by the processor, carry out the steps of the method for a similar target person identification process based on funds transaction relationship data according to any of claims 6 to 9.
12. A computer-readable storage medium having stored thereon a computer program executable by a processor to perform the steps of the method of any one of claims 6 to 9 for a similar target person identification process based on funds transaction relationship data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110940032.XA CN113641906A (en) | 2021-08-16 | 2021-08-16 | System, method, device, processor and medium for realizing similar target person identification processing based on fund transaction relation data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110940032.XA CN113641906A (en) | 2021-08-16 | 2021-08-16 | System, method, device, processor and medium for realizing similar target person identification processing based on fund transaction relation data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113641906A true CN113641906A (en) | 2021-11-12 |
Family
ID=78422311
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110940032.XA Pending CN113641906A (en) | 2021-08-16 | 2021-08-16 | System, method, device, processor and medium for realizing similar target person identification processing based on fund transaction relation data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113641906A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114741433A (en) * | 2022-06-09 | 2022-07-12 | 北京芯盾时代科技有限公司 | Community mining method, device, equipment and storage medium |
| CN115048554A (en) * | 2022-05-26 | 2022-09-13 | 支付宝(杭州)信息技术有限公司 | Graph data analysis method, device, equipment, medium and program product |
| CN115964549A (en) * | 2023-03-03 | 2023-04-14 | 北京芯盾时代科技有限公司 | Community mining method, device, equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110807103A (en) * | 2019-10-18 | 2020-02-18 | 中国银联股份有限公司 | Knowledge graph construction method and device, electronic equipment and storage medium |
| CN112199600A (en) * | 2020-10-29 | 2021-01-08 | 腾讯科技（深圳）有限公司 | Target object identification method and device |
| CN112465637A (en) * | 2020-11-05 | 2021-03-09 | 中国人民大学 | Anti-money laundering intelligent automatic sensing method and computer readable storage medium |
-
2021
- 2021-08-16 CN CN202110940032.XA patent/CN113641906A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110807103A (en) * | 2019-10-18 | 2020-02-18 | 中国银联股份有限公司 | Knowledge graph construction method and device, electronic equipment and storage medium |
| CN112199600A (en) * | 2020-10-29 | 2021-01-08 | 腾讯科技（深圳）有限公司 | Target object identification method and device |
| CN112465637A (en) * | 2020-11-05 | 2021-03-09 | 中国人民大学 | Anti-money laundering intelligent automatic sensing method and computer readable storage medium |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115048554A (en) * | 2022-05-26 | 2022-09-13 | 支付宝(杭州)信息技术有限公司 | Graph data analysis method, device, equipment, medium and program product |
| CN114741433A (en) * | 2022-06-09 | 2022-07-12 | 北京芯盾时代科技有限公司 | Community mining method, device, equipment and storage medium |
| CN115964549A (en) * | 2023-03-03 | 2023-04-14 | 北京芯盾时代科技有限公司 | Community mining method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110070141B (en) | Network intrusion detection method | |
| Mayer et al. | Exposing fake images with forensic similarity graphs | |
| CN113641906A (en) | System, method, device, processor and medium for realizing similar target person identification processing based on fund transaction relation data | |
| CN106203377B (en) | A kind of coal dust image-recognizing method | |
| Gill et al. | Fruit Image Classification Using Deep Learning. | |
| Salman et al. | Classification of real and fake human faces using deep learning | |
| Cuevas et al. | Automatic multiple circle detection based on artificial immune systems | |
| KR20200075114A (en) | System and Method for Matching Similarity between Image and Text | |
| Luo et al. | SFA: small faces attention face detector | |
| CN110245693B (en) | Key information infrastructure asset identification method combined with mixed random forest | |
| CN113392894A (en) | Cluster analysis method and system for multi-group mathematical data | |
| CN112884569A (en) | Credit assessment model training method, device and equipment | |
| CN114783021A (en) | Intelligent detection method, device, equipment and medium for wearing of mask | |
| Sudhakar et al. | Ada-SISE: adaptive semantic input sampling for efficient explanation of convolutional neural networks | |
| KR102039244B1 (en) | Data clustering method using firefly algorithm and the system thereof | |
| CN113158084B (en) | Method, device, computer equipment and storage medium for processing movement track data | |
| CN111352926A (en) | Data processing method, device, equipment and readable storage medium | |
| CN113343123A (en) | Training method and detection method for generating confrontation multiple relation graph network | |
| CN108830302B (en) | Image classification method, training method, classification prediction method and related device | |
| CN113111804B (en) | Face detection method and device, electronic equipment and storage medium | |
| CN113469816A (en) | Digital currency identification method, system and storage medium based on multigroup technology | |
| CN106156785A (en) | Method for checking object and body detection device | |
| CN114862588A (en) | Block chain transaction behavior-oriented anomaly detection method | |
| CN111382628B (en) | Method and device for judging peer | |
| CN112422505A (en) | Network malicious traffic identification method based on high-dimensional extended key feature vector |
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 |