CN109872232A - It is related to illicit gain to legalize account-classification method, device, computer equipment and the storage medium of behavior - Google Patents
It is related to illicit gain to legalize account-classification method, device, computer equipment and the storage medium of behavior Download PDFInfo
- Publication number
- CN109872232A CN109872232A CN201910008122.8A CN201910008122A CN109872232A CN 109872232 A CN109872232 A CN 109872232A CN 201910008122 A CN201910008122 A CN 201910008122A CN 109872232 A CN109872232 A CN 109872232A
- Authority
- CN
- China
- Prior art keywords
- trading
- clique
- transaction
- account
- money laundering
- 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 58
- 238000004900 laundering Methods 0.000 claims abstract description 163
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 47
- 238000004590 computer program Methods 0.000 claims description 23
- 238000004458 analytical method Methods 0.000 claims description 17
- 238000012549 training Methods 0.000 claims description 16
- 238000012545 processing Methods 0.000 claims description 15
- 230000006870 function Effects 0.000 claims description 13
- 238000013527 convolutional neural network Methods 0.000 claims description 10
- 230000006399 behavior Effects 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000003252 repetitive effect Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 13
- 239000000203 mixture Substances 0.000 description 8
- 238000012546 transfer Methods 0.000 description 8
- 239000006185 dispersion Substances 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 239000012141 concentrate Substances 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 2
- 239000010931 gold Substances 0.000 description 2
- 229910052737 gold Inorganic materials 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000006116 polymerization reaction Methods 0.000 description 2
- 238000013145 classification model Methods 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000005406 washing Methods 0.000 description 1
Landscapes
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The embodiment of the invention discloses a kind of illicit gains that is related to legalize account-classification method, device, computer equipment and the storage medium of behavior, wherein the method includes being identified the historical trading detail of several trading accounts with determination corresponding money laundering suspicion clique according to preset rules；Determine that the transaction feature of each trading account in money laundering suspicion clique, the transaction feature include at least total transaction amount and trading frequency；Clustering is carried out to money laundering suspicion clique according to identified transaction feature and Agglomerative Hierarchical Clustering algorithm, to obtain the classification collection of preset quantity, wherein each classification collection includes a number of Transaction Account number.The present invention can be analyzed by data and carry out accurately mechanized classification to money laundering account, improved the working efficiency of account classification, reduced repetitive operation.
Description
Technical field
The present invention relates to data processing fields more particularly to a kind of illicit gain that is related to legalize the account classification side of behavior
Method, device, computer equipment and storage medium.
Background technique
The behavior that usual people legalize illicit gain is known as money laundering, and will be related to illicit gain and legalize the account of behavior
Regard as money laundering account in family.Money laundering not only seriously endangers the national safety of financial system and the stabilization of social economic order, but also
Financing is provided for terrorism, jeopardizes the security of the lives and property of people.Anti money washing account divides main with the work of identification at present
By the professional knowledge and process experience of analysis personnel, building suspect's transaction feature is analyzed one by one manually.This analysis
Work very labor intensive and time resource, not only efficiency is lower, but also can neglect certain important features；Shortage can be specific
The index for quantifying suspect's trading activity similitude causes analysis to lack precision；In new clique's account identification process, nothing
Method automatic processing can generate a large amount of inefficient duplicate work.
Summary of the invention
The embodiment of the present invention provides a kind of money laundering account-classification method, device, computer equipment and storage medium, Neng Goutong
It crosses data analysis and accurately mechanized classification is carried out to money laundering account, improve the working efficiency of account classification, reduce repetition
Sex work.
In a first aspect, the embodiment of the invention provides a kind of money laundering account-classification methods, this method comprises:
Identify the historical trading detail of several trading accounts with determination corresponding money laundering suspicion clique according to preset rules；
Determine the transaction feature of each trading account in money laundering suspicion clique, the transaction feature, which includes at least, to be handed over
Easy total value and trading frequency；
Clustering is carried out to money laundering suspicion clique according to identified transaction feature and Agglomerative Hierarchical Clustering algorithm, with
The classification collection of preset quantity is obtained, wherein each classification collection includes a number of trading account.
Second aspect, the embodiment of the invention also provides a kind of money laundering account classification device, which includes:
Clique's determination unit determines corresponding for identifying the historical trading detail of several trading accounts according to preset rules
Money laundering suspicion clique；
Characteristics determining unit, it is described for determining the transaction feature of each trading account in money laundering suspicion clique
Transaction feature includes at least total transaction amount and trading frequency；
Processing unit, for according to identified transaction feature and Agglomerative Hierarchical Clustering algorithm to money laundering suspicion clique into
Row clustering, to obtain the classification collection of preset quantity, wherein each classification collection includes a number of trading account.
The third aspect, the embodiment of the invention also provides a kind of computer equipments comprising memory and processor, it is described
Computer program is stored on memory, the processor realizes the above method when executing the computer program.
Fourth aspect, the embodiment of the invention also provides a kind of computer readable storage medium, the storage medium storage
There is computer program, the computer program can realize the above method when being executed by a processor.
The embodiment of the invention provides a kind of money laundering account-classification method, device, computer equipment and storage mediums.Its
In, which comprises identify that the historical trading detail of several trading accounts is disliked with the corresponding money laundering of determination according to preset rules
Doubts and suspicions partner；Determine the transaction feature of each trading account in money laundering suspicion clique, the transaction feature, which includes at least, to be handed over
Easy total value and trading frequency；Money laundering suspicion clique is carried out according to identified transaction feature and Agglomerative Hierarchical Clustering algorithm
Clustering, to obtain the classification collection of preset quantity, wherein each classification collection includes a number of trading account.The present invention
Embodiment carries out accurately mechanized classification to money laundering account, it can be achieved that improving account classification due to that can analyze by data
Working efficiency reduces the effect of repetitive operation.
Detailed description of the invention
Technical solution in order to illustrate the embodiments of the present invention more clearly, below will be to needed in embodiment description
Attached drawing is briefly described, it should be apparent that, drawings in the following description are some embodiments of the invention, general for this field
For logical technical staff, without creative efforts, it is also possible to obtain other drawings based on these drawings.
Fig. 1 is a kind of flow diagram of money laundering account-classification method provided in an embodiment of the present invention；
Fig. 1 a is a kind of application scenarios schematic diagram of money laundering account-classification method provided in an embodiment of the present invention；
Fig. 2 is a kind of sub-process schematic diagram of money laundering account-classification method provided in an embodiment of the present invention；
Fig. 3 is a kind of sub-process schematic diagram of money laundering account-classification method provided in an embodiment of the present invention；
Fig. 4 be another embodiment of the present invention provides a kind of money laundering account-classification method flow diagram；
Fig. 5 is a kind of schematic block diagram of money laundering account classification device provided in an embodiment of the present invention；
Fig. 6 is a kind of schematic frame of clique's determination unit of money laundering account classification device provided in an embodiment of the present invention
Figure；
Fig. 7 is a kind of schematic block diagram of the processing unit of money laundering account classification device provided in an embodiment of the present invention；
Fig. 8 be another embodiment of the present invention provides a kind of money laundering account classification device schematic block diagram；
Fig. 9 is a kind of computer equipment structure composition schematic diagram provided in an embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on this hair
Embodiment in bright, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall within the protection scope of the present invention.
It should be appreciated that ought use in this specification and in the appended claims, term " includes " and "comprising" instruction
Described feature, entirety, step, operation, the presence of element and/or component, but one or more of the other feature, whole is not precluded
Body, step, operation, the presence or addition of element, component and/or its set.
It is also understood that mesh of the term used in this description of the invention merely for the sake of description specific embodiment
And be not intended to limit the present invention.As description of the invention and it is used in the attached claims, unless on
Other situations are hereafter clearly indicated, otherwise " one " of singular, "one" and "the" are intended to include plural form.
Fig. 1 and Fig. 1 a is please referred to, Fig. 1 is a kind of exemplary flow of money laundering account-classification method provided by the embodiments of the present application
Figure, Fig. 1 a is the schematic diagram of a scenario of money laundering account-classification method in the embodiment of the present application.The money laundering account-classification method is applied to
In management server 10.Management server 10 can analyze the friendship of multiple trading accounts in transaction system 20 according to preset rules
Easy detail is so that it is determined that money laundering suspicion clique, determines each friendship further according to the transaction details of each trading account in suspicion clique
The transaction feature of easy account, and transaction feature and Euclidean distance algorithm, cluster hierarchical algorithms is combined to carry out suspect's account
Accurate classification.Each step of the money laundering account-classification method will be introduced in detail with the angle of management server 10 below.
Referring to Fig. 1, Fig. 1 is a kind of schematic flow diagram of money laundering account-classification method provided in an embodiment of the present invention.Such as
Shown in Fig. 1, the step of this method includes step S101~S103.
Step S101 identifies that the historical trading detail of several trading accounts is disliked with the corresponding money laundering of determination according to preset rules
Doubts and suspicions partner.
In the present embodiment, the historical trading of several trading accounts in the available transaction system of management server is bright
Carefully, wherein transaction system can be bank and other financial mechanism, and trading account is to handle used in transaction agent in transaction system
The account of financial business.Each trading account can include a plurality of historical trading detail, and it is right that historical trading detail can be its
The a series of operation that the trading account answered carries out in transaction system, can specifically include such as transaction amount, trading frequency, friendship
The easy relevant informations such as type and counterparty.In order to determine money laundering suspicion clique, need according to preset rules to several transaction
The historical trading detail of account is analyzed.The preset rules refer to preset for determining money laundering suspicion clique
Rule.In addition, may include multiple trading accounts in the money laundering suspicion clique determined, trading account therein can be considered
Money laundering account can carry out classification division to the trading account in money laundering suspicion clique by the analysis of following step.
As another embodiment, as shown in Fig. 2, the step S101 can specifically include step S201~S203.
Step S201 obtains the historical trading detail of several trading accounts.Wherein, management server can be from transaction system
The historical trading detail for obtaining several trading accounts can determine each trading account according to the analysis to historical trading detail
Property.
Step S202 is analyzed multiple with determination according to historical trading detail of the preset clique's recognizer to acquisition
Trade clique, wherein each transaction clique includes multiple trading accounts.Wherein, preset clique's recognizer refers in advance
What is be arranged can go out respectively the algorithm of multiple transaction cliques according to the relevant information of existing historical trading detail.It will be all
Trading account is analyzed by above-mentioned clique's recognizer, so that multiple transaction cliques are obtained, wherein each transaction clique
In trading account between have between certain relevance and each trading account there are certain intersection behavior, such as both
Between used a transaction IP, or converged money etc. to the same counterparty.
Step S203 classifies to the transaction clique according to default two disaggregated models, to determine money laundering suspicion clique.
Wherein, preset two disaggregated models refer to it is pre-set can judge trade clique whether be money laundering suspicion clique model.I.e.
After the clique that will trade inputs default two disaggregated model, can judge whether it is money laundering suspicion clique.Default two classification
Model can be the model as obtained from being trained to neural network.
As further embodiment, before step S203, further includes:
Step S203a, by preset sample set training convolutional neural networks to obtain money laundering suspicion clique for identification
Default two disaggregated models.Wherein, preset sample set is the pre-set sample for training convolutional neural networks.It is described
Preset sample set includes the verifying collection for the training set of training convolutional neural networks and for being verified, convolutional Neural
Network can obtain default two disaggregated models under the common training of training set and verifying collection, in order to identify trade clique whether be
Money laundering suspicion clique.
As another embodiment, the step S101 be can specifically include: be based on suspicious degree function and letter according to preset
Corporations' recognizer of breath entropy identifies the historical trading detail of several trading accounts with determination corresponding money laundering suspicion clique.
Specifically, the corresponding bank card of each trading account, it can be using each bank card as composition trade network
A node then one line of their corresponding nodes is connected when there are direct money transfer transactions between two bank cards,
To constitute the side of trade network.The corresponding bank card of All Activity account and the money transfer transactions between them are converted into accordingly
The side of trade network node and trade network, to obtain a financial transaction network.It, can be with according to above-mentioned financial transaction network
For the suspicious degree function of each node definition, comentropy is defined for corporations, then determines the society based on suspicious degree function and comentropy
Group's recognizer.It can be efficiently identified out according to corporations' recognizer based on suspicious degree function and comentropy and be hidden in finance
Money laundering suspicion clique in trade network, while we are facilitated to the further analysis of the internal structure of clique and is further excavated
In clique between different nodes fund flowing relation.
Step S102 determines the transaction feature of each trading account in money laundering suspicion clique, the transaction feature
Including at least total transaction amount and trading frequency.
In the present embodiment, due to including multiple trading accounts in money laundering suspicion clique, and each trading account includes
Several transaction details, therefore can determine the transaction feature of each trading account according to all transaction details.The transaction is special
Sign may include the key data that can embody trading account as money laundering account, such as the number such as total transaction amount and trading frequency
According to.Wherein total transaction amount refer to trading account within the scope of certain time the amount of money transferred accounts of oriented trading object it is total
Number.And trading frequency refers to the number that trading account is transferred accounts within the scope of certain time, shows the transaction account if transaction frequently
Status of the family in money laundering suspicion clique is more important.
In addition, the transaction feature can also include type of transaction, transfer accounts number and friendship with same transaction number
The characteristic points such as easy mode, transactions velocity and test transaction.Wherein, type of transaction may include that Internetbank is transferred accounts or cash transfer etc.
The mode transferred accounts.Number of transferring accounts with same transaction amount refers to the identical number of the amount of money that trading account produces.And it trades
Mode may include that fund dispersion is transferred to concentration and produces, concentrates and be transferred to dispersion and produce, disperse to be transferred to dispersion and produce；Transactions velocity is then
Refer to that same fund is transferred to the speed produced in the account.Test transaction refers to open trading account after, appearance it is a large amount of
There is a large amount of block trade again later in penny ante.Pass through the knowledge of the two important features of total transaction amount and trading frequency
Not and the features such as type of transaction assist in identifying, and can further determine between each account in money laundering suspicion clique
Classification division.
Step S103 gathers money laundering suspicion clique according to identified transaction feature and Agglomerative Hierarchical Clustering algorithm
Class divides, to obtain the classification collection of preset quantity, wherein each classification collection includes a number of trading account.
In the present embodiment, hierarchical clustering be exactly in layer cluster, wherein from bottom to top to small classification into
Row polymerization, is called Agglomerative Hierarchical Clustering algorithm.Each sample point is regarded into a class cluster when it refers specifically to initial, so former
The size of beginning class cluster is equal to the number of sample point, then merges these initial class clusters according to certain criterion, until reaching certain
Condition or the classification number for reaching setting.Therefore according to the transaction feature of identified each trading account, money laundering can be disliked
Trading account in doubts and suspicions partner carries out clustering, and reaches the classification collection of preset quantity, and each classification collection representative belongs to same
The trading account of class.Therefore classified using trading account of the Agglomerative Hierarchical Clustering algorithm to money laundering suspicion clique, similarity degree
High trading account is easier to be gathered in identical classification, the classification results of coacervate hierarchical clustering and other clustering methods
Compared to more explanatory.
For example, cluster result can substantially be divided into when preset quantity is three classification collection: the first kind is on a large scale and solid
Determine the people that opponent trades；Second class is that large quantities of funds are distributed to the people of numerous accounts；Third class is largely to carry out small rule
Mould transaction and the identical people of transaction amount.Accordingly, above-mentioned first kind people can be money laundering main body；The artificial go-between of second class,
The fund of money laundering main body is distributed to regather after individual and returns main body back；Third class can be casual household, and number is numerous,
Main body is returned to by go-between after respectively microfinance can be cleaned.It can be seen that referring to above-mentioned cluster result, it can be to money laundering
Suspicion clique is accurately divided, and identifies division of labor situation of all kinds of suspects in crime of laundering behavior in clique.
As another embodiment, as shown in figure 3, the step S103 may include step S301~S303.
Step S301, according to identified transaction feature and Euclidean distance algorithm to the transaction account in money laundering suspicion clique
Euclidean distance between family is calculated two-by-two, to obtain Euclidean distance identical with the quantity of trading account.
Wherein, Euclidean distance algorithm is the algorithm for calculating user's similarity, can be according to commenting between two users jointly
The feature of valence is dimension, establishes the space of a multidimensional, and by user to the coordinate system of the evaluation of estimate composition in single dimension
Position of the user in this various dimensions space can be positioned, to calculate the distance between two positions, which makees
It can reflect the similarity degree between two users for Euclidean distance.Therefore using the transaction feature of identified trading account as dimension
Degree can construct hyperspace, to calculate the Euclidean distance between two different trading accounts.
Step S302 carries out clustering to obtained Euclidean distance according to Agglomerative Hierarchical Clustering algorithm, determines Euclidean
Apart from the smallest two trading accounts.
Wherein, after being analyzed by Agglomerative Hierarchical Clustering algorithm all Euclidean distances, can determine it is European away from
From the smallest two trading accounts, at the same can using above-mentioned two trading account as one kind, and using remaining trading account as
It is another kind of, to carry out the classification of next step.
Step S303, by the median of the transaction feature for the described two trading accounts being calculated and money laundering suspicion clique
In remaining trading account carry out the calculating of Euclidean distance, and according to the preset quantity of cluster and Agglomerative Hierarchical Clustering algorithm into
Row circulation clustering, to obtain the classification collection of preset quantity.
Wherein, it in order to which the trading account in money laundering suspicion clique to be categorized into the classification collection of preset quantity, needs to count
The median of the transaction feature of described two trading accounts is calculated, to determine the median with its in money laundering suspicion clique again
Euclidean distance between his trading account, and clustering is carried out again by Agglomerative Hierarchical Clustering algorithm, so recycle, until
Obtain the classification collection of preset quantity.
To sum up, the embodiment of the present invention carries out accurately mechanized classification to money laundering account due to that can analyze by data,
The working efficiency that raising account classification can be achieved, reduces the effect of repetitive operation.
Referring to Fig. 4, Fig. 4 be another embodiment of the present invention provides a kind of money laundering account-classification method exemplary flow
Figure.As shown in figure 4, the step of this method includes step S401~S407.In the present embodiment, the preset quantity is three classes,
The classification of the trading account includes money laundering main body, go-between and casual household, wherein with the step S101-S103 in above-described embodiment
The relevant explanation of similar step and it is described in detail that details are not described herein, the following detailed description of being increased in the present embodiment
Step.
Step S401 identifies that the historical trading detail of several trading accounts is disliked with the corresponding money laundering of determination according to preset rules
Doubts and suspicions partner.
Step S402 determines the transaction feature of each trading account in money laundering suspicion clique, the transaction feature
Including at least total transaction amount and trading frequency.
Step S403 gathers money laundering suspicion clique according to identified transaction feature and Agglomerative Hierarchical Clustering algorithm
Class divides, to obtain the classification collection of preset quantity, wherein each classification collection includes a number of trading account.
Step S404 determines that the transaction amount for the trading account that each classification is concentrated is corresponding according to preset scoring criterion
Score value and the corresponding score value of trading frequency.
In the present embodiment, in order to further determine the corresponding classification collection of money laundering theme, the corresponding classification collection of go-between with
And classification collection corresponding to casual household, the trade gold for the trading account for needing to determine that each classification is concentrated according to preset scoring criterion
The corresponding score value of volume and the corresponding score value of trading frequency.Wherein preset scoring criterion refers to preset according to user
The scoring criterion about transaction amount and trading frequency for needing to be arranged, for example, being preset when transaction amount is 10 ten thousand to 20 ten thousand
Scoring be 60 points, when transaction amount be 20 ten thousand to 50 ten thousand when, it is preset scoring be 70 points, and so on.There are column such as, when
Trading frequency is the transaction count within one week when being within 2 times, and preset scoring is 50 points, when trading frequency is within one week
Transaction count be 3 times to 8 times when, it is preset scoring be 60 points, similarly, can once analogize.
Step S405 calculates the transaction account that each classification is concentrated according to the default weight ratio of transaction amount and trading frequency
The corresponding total weight score value in family.
In the present embodiment, the transaction feature for the trading account concentrated for each classification of comprehensive analysis, it is also necessary in advance
Default weight ratio about transaction amount and trading frequency is set, is arrived each according to what above-mentioned default Quan Chong Bi Eng can calculate
The corresponding total weight score value of trading account that classification is concentrated.
Step S406 calculates each classification and concentrates quantity of total weight score value more than the trading account of preset threshold and such
The accounting value of the quantity for the trading account not collected.
In the present embodiment, by calculating the trading account that total weight score value that each classification is concentrated is more than preset threshold
The accounting value of the quantity of quantity and the trading account of category collection can further determine the property of classification collection on the whole.
Step S407 is ranked up the accounting value of all categories collection from big to small, with the corresponding money laundering main body of determination, in
Between people and casual household.
In the present embodiment, ordinary circumstance, money laundering main body, go-between and casual household are on transaction amount and trading frequency
Shared score value be from high to low, therefore being ranked up from big to small to the accounting of all categories collection, so as to
Corresponding money laundering theme, go-between and casual household are determined, to realize the fine division to money laundering suspicion clique.
Those having ordinary skill in the art is understood that realize all or part of the process in above-described embodiment method, is that can lead to
Computer program is crossed to instruct relevant hardware and complete, the program can be stored in a computer-readable storage medium
In, the program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein, the storage medium can be magnetic
Dish, CD, read-only memory (Read-Only Memory, ROM) etc..
Referring to Fig. 5, a kind of corresponding above-mentioned money laundering account-classification method, the embodiment of the present invention also propose a kind of money laundering account
Sorter, the device 100 include: clique's determination unit 101, characteristics determining unit 102 and processing unit 103.
Clique's determination unit 101, for identified according to preset rules the historical trading detail of several trading accounts with
Determine corresponding money laundering suspicion clique.
In the present embodiment, the historical trading of several trading accounts in the available transaction system of management server is bright
Carefully, wherein transaction system can be bank and other financial mechanism, and trading account is to handle used in transaction agent in transaction system
The account of financial business.Each trading account can include a plurality of historical trading detail, and it is right that historical trading detail can be its
The a series of operation that the trading account answered carries out in transaction system, can specifically include such as transaction amount, trading frequency, friendship
The easy relevant informations such as type and counterparty.In order to determine money laundering suspicion clique, need according to preset rules to several transaction
The historical trading detail of account is analyzed.The preset rules refer to preset for determining money laundering suspicion clique
Rule.In addition, may include multiple trading accounts in the money laundering suspicion clique determined, trading account therein can be considered
Money laundering account can carry out classification division to the trading account in money laundering suspicion clique by the analysis of following step.
As another embodiment, as shown in fig. 6, clique's determination unit 101 can specifically include detail acquiring unit
201, detail analytical unit 202 and clique's taxon 203.
The detail acquiring unit 201, for obtaining the historical trading detail of several trading accounts.Wherein, management service
Device can obtain the historical trading detail of several trading accounts from transaction system, can according to the analysis to historical trading detail
Determine the property of each trading account.
The detail analytical unit 202, for according to preset clique's recognizer to the historical trading detail of acquisition into
Row analysis is with the multiple transaction cliques of determination, wherein each transaction clique includes multiple trading accounts.Wherein, preset clique
Recognizer, which refers to, pre-set can go out respectively multiple transaction cliques according to the relevant information of existing historical trading detail
Algorithm.Trading account that will be all is analyzed by above-mentioned clique's recognizer, so that multiple transaction cliques are obtained,
Wherein with there are certain friendships between certain relevance and each trading account between the trading account in each transaction clique
Collection behavior, for example used a transaction IP between the two, or money etc. was converged to the same counterparty.
Clique's taxon 203, for being classified according to default two disaggregated models to the transaction clique, with true
Determine money laundering suspicion clique.Wherein, preset two disaggregated models refer to it is pre-set can judge trade clique whether be money laundering dislike
The model of doubts and suspicions partner.I.e. after the clique that will trade inputs default two disaggregated model, can judge whether it is money laundering suspicion
Clique.Default two disaggregated models can be the model as obtained from being trained to neural network.
As further embodiment, clique's determination unit 101 further include:
Training unit 203a dislikes for obtaining money laundering for identification by preset sample set training convolutional neural networks
Default two disaggregated models of doubts and suspicions partner.Wherein, preset sample set is the pre-set sample for training convolutional neural networks
This.The preset sample set includes the verifying collection for the training set of training convolutional neural networks and for being verified,
Convolutional neural networks can obtain default two disaggregated models under the common training of training set and verifying collection, in order to identify transaction group
Whether partner is money laundering suspicion clique.
As another embodiment, clique's determination unit 101 specifically can be used for according to preset based on suspicious degree letter
Several and comentropy corporations' recognizer identifies the historical trading detail of several trading accounts with determination corresponding money laundering suspicion group
Group.
Specifically, the corresponding bank card of each trading account, it can be using each bank card as composition trade network
A node then one line of their corresponding nodes is connected when there are direct money transfer transactions between two bank cards,
To constitute the side of trade network.The corresponding bank card of All Activity account and the money transfer transactions between them are converted into accordingly
The side of trade network node and trade network, to obtain a financial transaction network.It, can be with according to above-mentioned financial transaction network
For the suspicious degree function of each node definition, comentropy is defined for corporations, then determines the society based on suspicious degree function and comentropy
Group's recognizer.It can be efficiently identified out according to corporations' recognizer based on suspicious degree function and comentropy and be hidden in finance
Money laundering suspicion clique in trade network, while we are facilitated to the further analysis of the internal structure of clique and is further excavated
In clique between different nodes fund flowing relation.
The characteristics determining unit 102, the transaction for determining each trading account in money laundering suspicion clique are special
Sign, the transaction feature include at least total transaction amount and trading frequency.
In the present embodiment, due to including multiple trading accounts in money laundering suspicion clique, and each trading account includes
Several transaction details, therefore can determine the transaction feature of each trading account according to all transaction details.The transaction is special
Sign may include the key data that can embody trading account as money laundering account, such as the number such as total transaction amount and trading frequency
According to.Wherein total transaction amount refer to trading account within the scope of certain time the amount of money transferred accounts of oriented trading object it is total
Number.And trading frequency refers to the number that trading account is transferred accounts within the scope of certain time, shows the transaction account if transaction frequently
Status of the family in money laundering suspicion clique is more important.
In addition, the transaction feature can also include type of transaction, transfer accounts number and friendship with same transaction number
The characteristic points such as easy mode, transactions velocity and test transaction.Wherein, type of transaction may include that Internetbank is transferred accounts or cash transfer etc.
The mode transferred accounts.Number of transferring accounts with same transaction amount refers to the identical number of the amount of money that trading account produces.And it trades
Mode may include that fund dispersion is transferred to concentration and produces, concentrates and be transferred to dispersion and produce, disperse to be transferred to dispersion and produce；Transactions velocity is then
Refer to that same fund is transferred to the speed produced in the account.Test transaction refers to open trading account after, appearance it is a large amount of
There is a large amount of block trade again later in penny ante.Pass through the knowledge of the two important features of total transaction amount and trading frequency
Not and the features such as type of transaction assist in identifying, and can further determine between each account in money laundering suspicion clique
Classification division.
The processing unit 103, for being disliked according to identified transaction feature and Agglomerative Hierarchical Clustering algorithm to money laundering
Doubts and suspicions partner carries out clustering, to obtain the classification collection of preset quantity, wherein each classification collection includes a number of transaction
Account.
In the present embodiment, hierarchical clustering be exactly in layer cluster, wherein from bottom to top to small classification into
Row polymerization, is called Agglomerative Hierarchical Clustering algorithm.Each sample point is regarded into a class cluster when it refers specifically to initial, so former
The size of beginning class cluster is equal to the number of sample point, then merges these initial class clusters according to certain criterion, until reaching certain
Condition or the classification number for reaching setting.Therefore according to the transaction feature of identified each trading account, money laundering can be disliked
Trading account in doubts and suspicions partner carries out clustering, and reaches the classification collection of preset quantity, and each classification collection representative belongs to same
The trading account of class.Therefore classified using trading account of the Agglomerative Hierarchical Clustering algorithm to money laundering suspicion clique, similarity degree
High trading account is easier to be gathered in identical classification, the classification results of coacervate hierarchical clustering and other clustering methods
Compared to more explanatory.
As another embodiment, as shown in fig. 7, the processing unit 103 may include metrics calculation unit 301, cluster
Analytical unit 302 and cycle analysis unit 303.
The metrics calculation unit 301, for being disliked according to identified transaction feature and Euclidean distance algorithm to money laundering
The Euclidean distance between trading account in doubts and suspicions partner is calculated two-by-two, identical with the quantity of trading account European to obtain
Distance.
Wherein, Euclidean distance algorithm is the algorithm for calculating user's similarity, can be according to commenting between two users jointly
The feature of valence is dimension, establishes the space of a multidimensional, and by user to the coordinate system of the evaluation of estimate composition in single dimension
Position of the user in this various dimensions space can be positioned, to calculate the distance between two positions, which makees
It can reflect the similarity degree between two users for Euclidean distance.Therefore using the transaction feature of identified trading account as dimension
Degree can construct hyperspace, to calculate the Euclidean distance between two different trading accounts.
The cluster analysis unit 302, for being gathered according to Agglomerative Hierarchical Clustering algorithm to obtained Euclidean distance
Alanysis determines the smallest two trading accounts of Euclidean distance.
Wherein, after being analyzed by Agglomerative Hierarchical Clustering algorithm all Euclidean distances, can determine it is European away from
From the smallest two trading accounts, at the same can using above-mentioned two trading account as one kind, and using remaining trading account as
It is another kind of, to carry out the classification of next step.
The cycle analysis unit 303, for by the centre of the transaction feature for the described two trading accounts being calculated
Value carries out the calculating of Euclidean distance with the remaining trading account in money laundering suspicion clique, and according to the preset quantity of cluster and coagulates
Poly- hierarchical clustering algorithm carries out circulation clustering, to obtain the classification collection of preset quantity.
Wherein, it in order to which the trading account in money laundering suspicion clique to be categorized into the classification collection of preset quantity, needs to count
The median of the transaction feature of described two trading accounts is calculated, to determine the median with its in money laundering suspicion clique again
Euclidean distance between his trading account, and clustering is carried out again by Agglomerative Hierarchical Clustering algorithm, so recycle, until
Obtain the classification collection of preset quantity.
Referring to Fig. 8, a kind of corresponding above-mentioned money laundering account-classification method, another embodiment of the present invention also propose a kind of money laundering
Account classification device, the device 400 include: that clique's determination unit 401, characteristics determining unit 402, processing unit 403, score value are true
Order member 404, weight calculation unit 405, accounting computing unit 406 and sequence determination unit 407.
Clique's determination unit 401, for identified according to preset rules the historical trading detail of several trading accounts with
Determine corresponding money laundering suspicion clique.
The characteristics determining unit 402, the transaction for determining each trading account in money laundering suspicion clique are special
Sign, the transaction feature include at least total transaction amount and trading frequency.
The processing unit 403, for being disliked according to identified transaction feature and Agglomerative Hierarchical Clustering algorithm to money laundering
Doubts and suspicions partner carries out clustering, to obtain the classification collection of preset quantity, wherein each classification collection includes a number of transaction
Account.
The score value determination unit 404, for determining the trading account of each classification concentration according to preset scoring criterion
The corresponding score value of transaction amount and the corresponding score value of trading frequency.
In the present embodiment, in order to further determine the corresponding classification collection of money laundering theme, the corresponding classification collection of go-between with
And classification collection corresponding to casual household, the trade gold for the trading account for needing to determine that each classification is concentrated according to preset scoring criterion
The corresponding score value of volume and the corresponding score value of trading frequency.Wherein preset scoring criterion refers to preset according to user
The scoring criterion about transaction amount and trading frequency for needing to be arranged, for example, being preset when transaction amount is 10 ten thousand to 20 ten thousand
Scoring be 60 points, when transaction amount be 20 ten thousand to 50 ten thousand when, it is preset scoring be 70 points, and so on.There are column such as, when
Trading frequency is the transaction count within one week when being within 2 times, and preset scoring is 50 points, when trading frequency is within one week
Transaction count be 3 times to 8 times when, it is preset scoring be 60 points, similarly, can once analogize.
The weight calculation unit 405, it is each for being calculated according to the default weight ratio of transaction amount and trading frequency
The corresponding total weight score value of trading account that classification is concentrated.
In the present embodiment, the transaction feature for the trading account concentrated for each classification of comprehensive analysis, it is also necessary in advance
Default weight ratio about transaction amount and trading frequency is set, is arrived each according to what above-mentioned default Quan Chong Bi Eng can calculate
The corresponding total weight score value of trading account that classification is concentrated.
The accounting computing unit 406 is more than the transaction of preset threshold for calculating each classification to concentrate total weight score value
The accounting value of the quantity of the trading account of the quantity and category collection of account.
In the present embodiment, by calculating the trading account that total weight score value that each classification is concentrated is more than preset threshold
The accounting value of the quantity of quantity and the trading account of category collection can further determine the property of classification collection on the whole.
The sequence determination unit 407, is ranked up, from big to small to determine phase for the accounting value to all categories collection
Money laundering main body, go-between and the casual household answered.
In the present embodiment, ordinary circumstance, money laundering main body, go-between and casual household are on transaction amount and trading frequency
Shared score value be from high to low, therefore being ranked up from big to small to the accounting of all categories collection, so as to
Corresponding money laundering theme, go-between and casual household are determined, to realize the fine division to money laundering suspicion clique.
It should be noted that it is apparent to those skilled in the art that, above-mentioned money laundering account classification device
100 and each unit specific implementation process, can with reference to the corresponding description in preceding method embodiment, for convenience of description and
Succinctly, details are not described herein.
As seen from the above, in hardware realization, above clique's determination unit 101, characteristics determining unit 102 and processing are single
Member 103 etc. can be embedded in the form of hardware or the device reported a case to the security authorities independently of life insurance in, can also be stored in and wash in a software form
In the memory of money account classification device, the corresponding operation of above each unit is executed so that processor calls.The processor can
Think central processing unit (CPU), microprocessor, single-chip microcontroller etc..
Above-mentioned money laundering account classification device can be implemented as a kind of form of computer program, and computer program can be such as
It is run in computer equipment shown in Fig. 9.
Fig. 9 is a kind of structure composition schematic diagram of computer equipment of the present invention.The equipment can be server, wherein clothes
Business device can be independent server, be also possible to the server cluster of multiple server compositions.
Referring to Fig. 9, which includes processor 502, memory, the memory connected by system bus 501
Reservoir 504 and network interface 505, wherein memory may include non-volatile memory medium 503 and built-in storage 504.
The non-volatile memory medium 503 can storage program area 5031 and computer program 5032, the computer program
5032 are performed, and processor 502 may make to execute a kind of money laundering account-classification method.
The processor 502 supports the operation of entire computer equipment 500 for providing calculating and control ability.
The built-in storage 504 provides environment for the operation of the computer program 5032 in non-volatile memory medium 503, should
When computer program 5032 is executed by processor 502, processor 502 may make to execute a kind of money laundering account-classification method.
The network interface 505 is used to carry out network communication with other equipment.It will be understood by those skilled in the art that in Fig. 9
The structure shown, only the block diagram of part-structure relevant to application scheme, does not constitute and is applied to application scheme
The restriction of computer equipment 500 thereon, specific computer equipment 500 may include more more or fewer than as shown in the figure
Component perhaps combines certain components or with different component layouts.
Wherein, the processor 502 is for running computer program 5032 stored in memory, to realize following step
It is rapid: to identify the historical trading detail of several trading accounts with determination corresponding money laundering suspicion clique according to preset rules；Determine institute
The transaction feature of each trading account in money laundering suspicion clique is stated, the transaction feature includes at least total transaction amount and transaction
Frequency；Clustering is carried out to money laundering suspicion clique according to identified transaction feature and Agglomerative Hierarchical Clustering algorithm, with
To the classification collection of preset quantity, wherein each classification collection includes a number of trading account.
In one embodiment, processor 502 is realizing the history friendship that several trading accounts are identified according to preset rules
It the step of easy detail corresponding with determination money laundering suspicion clique, is implemented as follows step: obtaining the history of several trading accounts
Transaction details；It is analyzed according to historical trading detail of the preset clique's recognizer to acquisition with the multiple transaction groups of determination
Group, wherein each transaction clique includes multiple trading accounts；The transaction clique is divided according to default two disaggregated models
Class, to determine money laundering suspicion clique.
In one embodiment, processor 502 is preset two disaggregated models in the realization basis and is carried out to the transaction clique
Classification, the step of to determine money laundering suspicion clique before, be implemented as follows step: by preset sample set training convolutional mind
Through network to obtain default two disaggregated models of money laundering suspicion clique for identification.
In one embodiment, processor 502 is realizing the history friendship that several trading accounts are identified according to preset rules
The step of easy detail corresponding with determination money laundering suspicion clique, it is implemented as follows step: according to preset based on suspicious degree letter
Several and comentropy corporations' recognizer identifies the historical trading detail of several trading accounts with determination corresponding money laundering suspicion group
Group.
In one embodiment, processor 502 is realizing the transaction feature according to determined by and Agglomerative Hierarchical Clustering
Algorithm carries out clustering to money laundering suspicion clique and is implemented as follows step when obtaining the step of classification collection of preset quantity
It is rapid: according to identified transaction feature and Euclidean distance algorithm between the trading account in money laundering suspicion clique it is European away from
From being calculated two-by-two, to obtain Euclidean distance identical with the quantity of trading account；According to Agglomerative Hierarchical Clustering algorithm to institute
Obtained Euclidean distance carries out clustering, determines the smallest two trading accounts of Euclidean distance；Described two will be calculated
The median of the transaction feature of a trading account and the remaining trading account in money laundering suspicion clique carry out the calculating of Euclidean distance,
And circulation clustering is carried out according to the preset quantity of cluster and Agglomerative Hierarchical Clustering algorithm, to obtain the classification of preset quantity
Collection.
In one embodiment, the preset quantity is three classes, and the classification of the trading account includes money laundering main body, go-between
And casual household, processor 502 also realize following steps: determining the trading account that each classification is concentrated according to preset scoring criterion
The corresponding score value of transaction amount and the corresponding score value of trading frequency；According to the default weight ratio of transaction amount and trading frequency
Calculate the corresponding total weight score value of trading account that each classification is concentrated；It is more than default for calculating each classification and concentrating total weight score value
The accounting value of the quantity of the trading account of the quantity and category collection of the trading account of threshold values；To the accounting value of all categories collection from
Arrive greatly it is small be ranked up, with the corresponding money laundering main body of determination, go-between and casual household.
It should be appreciated that in the embodiment of the present application, processor 502 can be central processing unit (Central
Processing Unit, CPU), which can also be other general processors, digital signal processor (Digital
Signal Processor, DSP), specific integrated circuit (Application Specific Integrated Circuit,
ASIC), ready-made programmable gate array (Field-Programmable GateArray, FPGA) or other programmable logic devices
Part, discrete gate or transistor logic, discrete hardware components etc..Wherein, general processor can be microprocessor or
The processor is also possible to any conventional processor etc..
Those of ordinary skill in the art will appreciate that be realize above-described embodiment method in all or part of the process,
It is that relevant hardware can be instructed to complete by computer program.The computer program can be stored in a storage medium,
The storage medium is computer readable storage medium.The computer program is held by least one processor in the computer system
Row, to realize the process step of the embodiment of the above method.
Therefore, the present invention also provides a kind of storage mediums.The storage medium can be computer readable storage medium.This is deposited
Storage media is stored with computer program, which makes processor execute following steps when being executed by processor: according to pre-
If rule identifies the historical trading detail of several trading accounts with determination corresponding money laundering suspicion clique；Determine the money laundering suspicion
The transaction feature of each trading account in clique, the transaction feature include at least total transaction amount and trading frequency；According to
Identified transaction feature and Agglomerative Hierarchical Clustering algorithm carry out clustering to money laundering suspicion clique, to obtain preset quantity
Classification collection, wherein each classification collection includes a number of trading account.
In one embodiment, the processor is realized and described is identified according to preset rules executing the computer program
When the historical trading detail of several trading accounts is with the step of determination corresponding money laundering suspicion clique, it is implemented as follows step:
Obtain the historical trading detail of several trading accounts；It is carried out according to historical trading detail of the preset clique's recognizer to acquisition
Analysis is with the multiple transaction cliques of determination, wherein each transaction clique includes multiple trading accounts；According to default two disaggregated models
Classify to the transaction clique, to determine money laundering suspicion clique.
In one embodiment, the processor realizes the default two classification mould of the basis executing the computer program
Type classifies to the transaction clique, the step of to determine money laundering suspicion clique before, be implemented as follows step: by pre-
If sample set training convolutional neural networks to obtain default two disaggregated models of money laundering suspicion clique for identification.
In one embodiment, the processor is realized and described is identified according to preset rules executing the computer program
The step of historical trading detail of several trading accounts corresponding with determination money laundering suspicion clique, it is implemented as follows step: root
According to it is preset based on it is suspicious degree function and comentropy corporations' recognizer identify the historical trading detail of several trading accounts with
Determine corresponding money laundering suspicion clique.
In one embodiment, the processor realizes the transaction according to determined by executing the computer program
Feature and Agglomerative Hierarchical Clustering algorithm carry out clustering to money laundering suspicion clique, to obtain the step of the classification collection of preset quantity
When rapid, it is implemented as follows step: according to identified transaction feature and Euclidean distance algorithm in money laundering suspicion clique
Euclidean distance between trading account is calculated two-by-two, to obtain Euclidean distance identical with the quantity of trading account；According to
Agglomerative Hierarchical Clustering algorithm carries out clustering to obtained Euclidean distance, determines the smallest two transaction account of Euclidean distance
Family；By the remaining transaction account in the median of the transaction feature for the described two trading accounts being calculated and money laundering suspicion clique
Family carries out the calculating of Euclidean distance, and carries out circulation cluster point according to the preset quantity of cluster and Agglomerative Hierarchical Clustering algorithm
Analysis, to obtain the classification collection of preset quantity.
In one embodiment, the preset quantity is three classes, and the classification of the trading account includes money laundering main body, go-between
And casual household, the processor also execute the following steps: the trading account that each classification concentration is determined according to preset scoring criterion
The corresponding score value of transaction amount and the corresponding score value of trading frequency；According to the default weight of transaction amount and trading frequency
The corresponding total weight score value of trading account than calculating each classification concentration；It is more than pre- for calculating each classification and concentrating total weight score value
If the accounting value of the quantity of the quantity of the trading account of threshold values and the trading account of category collection；To the accounting value of all categories collection
It is ranked up from big to small, with the corresponding money laundering main body of determination, go-between and casual household.
The storage medium can be USB flash disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), magnetic disk
Or the various computer readable storage mediums that can store program code such as CD.
Those of ordinary skill in the art may be aware that list described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware
With the interchangeability of software, each exemplary composition and step are generally described according to function in the above description.This
A little functions are implemented in hardware or software actually, the specific application and design constraint depending on technical solution.Specially
Industry technical staff can use different methods to achieve the described function each specific application, but this realization is not
It is considered as beyond the scope of this invention.
In several embodiments provided by the present invention, it should be understood that disclosed device and method can pass through it
Its mode is realized.For example, the apparatus embodiments described above are merely exemplary.For example, the division of each unit, only
Only a kind of logical function partition, there may be another division manner in actual implementation.Such as multiple units or components can be tied
Another system is closed or is desirably integrated into, or some features can be ignored or not executed.
The steps in the embodiment of the present invention can be sequentially adjusted, merged and deleted according to actual needs.This hair
Unit in bright embodiment device can be combined, divided and deleted according to actual needs.In addition, in each implementation of the present invention
Each functional unit in example can integrate in one processing unit, is also possible to each unit and physically exists alone, can also be with
It is that two or more units are integrated in one unit.
If the integrated unit is realized in the form of SFU software functional unit and when sold or used as an independent product,
It can store in one storage medium.Based on this understanding, technical solution of the present invention is substantially in other words to existing skill
The all or part of part or the technical solution that art contributes can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, terminal or network equipment etc.) it performs all or part of the steps of the method described in the various embodiments of the present invention.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can readily occur in various equivalent modifications or replace
It changes, these modifications or substitutions should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with right
It is required that protection scope subject to.
Claims (10)
1. a kind of illicit gain that is related to legalizes the account-classification method of behavior, which is characterized in that the described method includes:
Identify the historical trading detail of several trading accounts with determination corresponding money laundering suspicion clique according to preset rules；
Determine the transaction feature of each trading account in money laundering suspicion clique, it is total that the transaction feature includes at least transaction
Volume and trading frequency；
Clustering is carried out to money laundering suspicion clique according to identified transaction feature and Agglomerative Hierarchical Clustering algorithm, to obtain
The classification collection of preset quantity, wherein each classification collection includes a number of trading account.
2. the method as described in claim 1, which is characterized in that the history for identifying several trading accounts according to preset rules
The step of transaction details corresponding with determination money laundering suspicion clique, comprising:
Obtain the historical trading detail of several trading accounts；
It is analyzed according to historical trading detail of the preset clique's recognizer to acquisition with the multiple transaction cliques of determination,
In, each transaction clique includes multiple trading accounts；
Classified according to default two disaggregated models to the transaction clique, to determine money laundering suspicion clique.
3. method according to claim 2, which is characterized in that the basis preset two disaggregated models to the transaction clique into
Row classification, the step of to determine money laundering suspicion clique before, further includes:
By preset sample set training convolutional neural networks to obtain the default two classification mould of money laundering suspicion clique for identification
Type.
4. the method as described in claim 1, which is characterized in that the history for identifying several trading accounts according to preset rules
The step of transaction details corresponding with determination money laundering suspicion clique, comprising:
The historical trading of several trading accounts is identified according to preset corporations' recognizer based on suspicious degree function and comentropy
Detail is with determination corresponding money laundering suspicion clique.
5. the method as described in claim 1, which is characterized in that the transaction feature according to determined by and cohesion level are poly-
Class algorithm carries out clustering to money laundering suspicion clique, the step of to obtain the classification collection of preset quantity, comprising:
According to identified transaction feature and Euclidean distance algorithm to European between the trading account in money laundering suspicion clique
Distance is calculated two-by-two, to obtain Euclidean distance identical with the quantity of trading account；
Clustering is carried out to obtained Euclidean distance according to Agglomerative Hierarchical Clustering algorithm, determines that Euclidean distance is two the smallest
Trading account；
By the remaining transaction in the median of the transaction feature for the described two trading accounts being calculated and money laundering suspicion clique
Account carries out the calculating of Euclidean distance, and carries out circulation cluster point according to the preset quantity of cluster and Agglomerative Hierarchical Clustering algorithm
Analysis, to obtain the classification collection of preset quantity.
6. the method as described in claim 1, which is characterized in that the preset quantity is three classes, the classification of the trading account
Including money laundering main body, go-between and casual household, the method also includes:
The corresponding score value of transaction amount and the transaction of the trading account that each classification is concentrated are determined according to preset scoring criterion
The corresponding score value of frequency；
The corresponding total power of trading account that each classification is concentrated is calculated according to the default weight ratio of transaction amount and trading frequency
Weight score value；
Calculating each classification and concentrating total weight score value is more than the quantity of the trading account of preset threshold and the transaction account of category collection
The accounting value of the quantity at family；
The accounting value of all categories collection is ranked up from big to small, with the corresponding money laundering main body of determination, go-between and casual household.
7. a kind of illicit gain that is related to legalizes the account classification device of behavior, which is characterized in that described device includes:
Clique's determination unit, for identifying that the historical trading detail of several trading accounts is washed accordingly with determination according to preset rules
Money suspicion clique；
Characteristics determining unit, for determining the transaction feature of each trading account in money laundering suspicion clique, the transaction
Feature includes at least total transaction amount and trading frequency；
Processing unit, for being gathered according to identified transaction feature and Agglomerative Hierarchical Clustering algorithm to money laundering suspicion clique
Class divides, to obtain the classification collection of preset quantity, wherein each classification collection includes a number of trading account.
8. device as claimed in claim 7, which is characterized in that clique's determination unit, comprising:
Detail acquiring unit, for obtaining the historical trading detail of several trading accounts；
Detail analytical unit, for being analyzed according to historical trading detail of the preset clique's recognizer to acquisition with determination
Multiple transaction cliques, wherein each transaction clique includes multiple trading accounts；
Clique's taxon, for being classified according to default two disaggregated models to the transaction clique, to determine money laundering suspicion
Clique.
9. a kind of computer equipment, which is characterized in that the computer equipment includes memory and processor, on the memory
It is stored with computer program, the processor is realized as described in any one of claim 1-6 when executing the computer program
Method.
10. a kind of computer readable storage medium, which is characterized in that the storage medium is stored with computer program, the meter
Calculation machine program makes the processor execute such as method of any of claims 1-6 when being executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910008122.8A CN109872232A (en) | 2019-01-04 | 2019-01-04 | It is related to illicit gain to legalize account-classification method, device, computer equipment and the storage medium of behavior |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910008122.8A CN109872232A (en) | 2019-01-04 | 2019-01-04 | It is related to illicit gain to legalize account-classification method, device, computer equipment and the storage medium of behavior |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109872232A true CN109872232A (en) | 2019-06-11 |
Family
ID=66917520
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910008122.8A Pending CN109872232A (en) | 2019-01-04 | 2019-01-04 | It is related to illicit gain to legalize account-classification method, device, computer equipment and the storage medium of behavior |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109872232A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110706111A (en) * | 2019-10-15 | 2020-01-17 | 支付宝(杭州)信息技术有限公司 | Method and device for identifying suspicious transaction account, storage medium and server |
| CN110827038A (en) * | 2019-10-11 | 2020-02-21 | 支付宝(杭州)信息技术有限公司 | Account authority management method and device |
| CN110852809A (en) * | 2019-11-15 | 2020-02-28 | 腾讯科技（深圳）有限公司 | Data processing method, device, equipment and medium |
| CN110895758A (en) * | 2019-12-02 | 2020-03-20 | 中国银行股份有限公司 | Screening method, device and system for credit card account with cheating transaction |
| CN111145027A (en) * | 2019-12-31 | 2020-05-12 | 众安信息技术服务有限公司 | Suspected money laundering transaction identification method and device |
| CN111242763A (en) * | 2020-01-07 | 2020-06-05 | 北京明略软件系统有限公司 | Method and device for determining target user group |
| CN112541765A (en) * | 2019-09-20 | 2021-03-23 | 京东数字科技控股有限公司 | Method and apparatus for detecting suspicious transactions |
| CN112950357A (en) * | 2021-03-22 | 2021-06-11 | 工银科技有限公司 | Transaction abnormal group partner identification method and device |
| CN113129018A (en) * | 2021-05-17 | 2021-07-16 | 无锡航吴科技有限公司 | Financing platform account classification method and system |
| CN113393244A (en) * | 2021-05-25 | 2021-09-14 | 中国银联股份有限公司 | Abnormal account identification method, device, equipment and medium |
| CN114742655A (en) * | 2022-06-13 | 2022-07-12 | 杭银消费金融股份有限公司 | Anti-money laundering behavior recognition system based on machine learning |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030033228A1 (en) * | 2000-11-30 | 2003-02-13 | Rowan Bosworth-Davies | Countermeasures for irregularities in financial transactions |
| CN104199832A (en) * | 2014-08-01 | 2014-12-10 | 西安理工大学 | Financial network unusual transaction community finding method based on information entropy |
| CN106156795A (en) * | 2016-07-05 | 2016-11-23 | 中国银联股份有限公司 | A kind of determination method and device of suspicious money laundering account |
| CN108228706A (en) * | 2017-11-23 | 2018-06-29 | 中国银联股份有限公司 | For identifying the method and apparatus of abnormal transaction corporations |
| CN108280755A (en) * | 2018-02-28 | 2018-07-13 | 阿里巴巴集团控股有限公司 | The recognition methods of suspicious money laundering clique and identification device |
-
2019
- 2019-01-04 CN CN201910008122.8A patent/CN109872232A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030033228A1 (en) * | 2000-11-30 | 2003-02-13 | Rowan Bosworth-Davies | Countermeasures for irregularities in financial transactions |
| CN104199832A (en) * | 2014-08-01 | 2014-12-10 | 西安理工大学 | Financial network unusual transaction community finding method based on information entropy |
| CN106156795A (en) * | 2016-07-05 | 2016-11-23 | 中国银联股份有限公司 | A kind of determination method and device of suspicious money laundering account |
| CN108228706A (en) * | 2017-11-23 | 2018-06-29 | 中国银联股份有限公司 | For identifying the method and apparatus of abnormal transaction corporations |
| CN108280755A (en) * | 2018-02-28 | 2018-07-13 | 阿里巴巴集团控股有限公司 | The recognition methods of suspicious money laundering clique and identification device |
Non-Patent Citations (1)
| Title |
|---|
| 陈敏: "认知计算导论", 30 April 2017, 华中科技大学出版社, pages: 76 - 81 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112541765A (en) * | 2019-09-20 | 2021-03-23 | 京东数字科技控股有限公司 | Method and apparatus for detecting suspicious transactions |
| CN110827038A (en) * | 2019-10-11 | 2020-02-21 | 支付宝(杭州)信息技术有限公司 | Account authority management method and device |
| CN110706111A (en) * | 2019-10-15 | 2020-01-17 | 支付宝(杭州)信息技术有限公司 | Method and device for identifying suspicious transaction account, storage medium and server |
| CN110706111B (en) * | 2019-10-15 | 2022-11-01 | 支付宝(杭州)信息技术有限公司 | Method and device for identifying suspicious transaction account, storage medium and server |
| CN110852809A (en) * | 2019-11-15 | 2020-02-28 | 腾讯科技（深圳）有限公司 | Data processing method, device, equipment and medium |
| CN110895758A (en) * | 2019-12-02 | 2020-03-20 | 中国银行股份有限公司 | Screening method, device and system for credit card account with cheating transaction |
| CN111145027A (en) * | 2019-12-31 | 2020-05-12 | 众安信息技术服务有限公司 | Suspected money laundering transaction identification method and device |
| CN111242763A (en) * | 2020-01-07 | 2020-06-05 | 北京明略软件系统有限公司 | Method and device for determining target user group |
| CN112950357A (en) * | 2021-03-22 | 2021-06-11 | 工银科技有限公司 | Transaction abnormal group partner identification method and device |
| CN112950357B (en) * | 2021-03-22 | 2024-03-15 | 工银科技有限公司 | Transaction abnormal group identification method and device |
| CN113129018A (en) * | 2021-05-17 | 2021-07-16 | 无锡航吴科技有限公司 | Financing platform account classification method and system |
| CN113129018B (en) * | 2021-05-17 | 2023-12-08 | 无锡航吴科技有限公司 | Financing platform account classification method and system |
| CN113393244A (en) * | 2021-05-25 | 2021-09-14 | 中国银联股份有限公司 | Abnormal account identification method, device, equipment and medium |
| CN114742655A (en) * | 2022-06-13 | 2022-07-12 | 杭银消费金融股份有限公司 | Anti-money laundering behavior recognition system based on machine learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109872232A (en) | It is related to illicit gain to legalize account-classification method, device, computer equipment and the storage medium of behavior | |
| Björkegren et al. | Behavior revealed in mobile phone usage predicts loan repayment | |
| CN109784636A (en) | Fraudulent user recognition methods, device, computer equipment and storage medium | |
| CN109242499A (en) | A kind of processing method of transaction risk prediction, apparatus and system | |
| CN109035003A (en) | Anti- fraud model modelling approach and anti-fraud monitoring method based on machine learning | |
| Shingi | A federated learning based approach for loan defaults prediction | |
| CN106022508A (en) | Method and apparatus for predicting user friend invitation behaviors of online financing platform | |
| CN106372938A (en) | Abnormal account identification method and system | |
| CN109615524B (en) | Money laundering crime group partner identification method, money laundering crime group partner identification device, computer equipment and storage medium | |
| CN108322473A (en) | User behavior analysis method and apparatus | |
| CN110135853A (en) | Clique's user identification method, device and equipment | |
| CN106156092B (en) | Data processing method and device | |
| CN106326913A (en) | Money laundering account determination method and device | |
| CN109670797A (en) | Pay route selecting method, apparatus, equipment and storage medium | |
| CN107886366A (en) | Generation method, sex fill method, terminal and the storage medium of Gender Classification model | |
| CN112488716B (en) | Abnormal event detection system | |
| CN106649517A (en) | Data mining method, device and system | |
| CN109818961A (en) | A kind of network inbreak detection method, device and equipment | |
| CN106408325A (en) | User consumption behavior prediction analysis method based on user payment information and system | |
| Crawford et al. | Dynamic selection of enumeration strategies for solving constraint satisfaction problems | |
| CN112836771A (en) | Business service point classification method and device, electronic equipment and storage medium | |
| Cao et al. | Bond rating using support vector machine | |
| CN112598225A (en) | Evaluation index determination method and apparatus, storage medium, and electronic apparatus | |
| CN108563786A (en) | Text classification and methods of exhibiting, device, computer equipment and storage medium | |
| Melian et al. | Customer churn prediction in telecommunication industry. A data analysis techniques approach |
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 |