CN114022272A

CN114022272A - Method and device for identifying hidden fund return

Info

Publication number: CN114022272A
Application number: CN202111325249.6A
Authority: CN
Inventors: 卢健
Original assignee: Industrial and Commercial Bank of China Ltd ICBC
Current assignee: Industrial and Commercial Bank of China Ltd ICBC
Priority date: 2021-11-10
Filing date: 2021-11-10
Publication date: 2022-02-08

Abstract

The invention provides a method and a device for identifying hidden fund return, which can be used in the financial field or other technical fields. The method comprises the following steps: according to first transaction data and full-amount customer transaction data after a target loan customer loans, the target loan customer is used as an initial node, a fund flow map of the target loan customer is constructed, customers participating in fund transaction are used as nodes in the fund flow map of the target loan customer, and fund transaction relations among the nodes are used as edges connecting the nodes; and identifying the recessive fund backflow information of the target loan customer according to the established fund flow map of the target loan customer, wherein the recessive fund backflow information is used for indicating whether the target loan customer has recessive fund backflow. The device is used for executing the method. The method and the device for identifying the hidden fund return flow provided by the embodiment of the invention can identify whether the loan customer has the hidden fund return flow.

Description

Method and device for identifying hidden fund return

Technical Field

The invention relates to the field of finance or other technologies, in particular to a method and a device for identifying hidden fund return.

Background

In recent years, financing products on banking lines are rapidly developed, and the traffic volume is increased year by year. However, some borrowers move loan funds between the borrower account, the counterparty account, and the third party customer account, appropriating the loan funds, and evading bank supervision. In the case of capital returns, "covert" capital returns generally have the following characteristics, as shown in FIG. 1:

the loan client A is credited to fund from the bank for some reasonable reason and then transfers to B, B again flows for a plurality of times through the bank account X, finally returns the fund to A from the bank, and A takes the fund and uses the fund for other unreasonable purposes.

The existing fund return recognition technology has no way to solve the above 'hidden' fund return case and has no complete solution.

Disclosure of Invention

In view of the problems in the prior art, embodiments of the present invention provide a method and an apparatus for identifying a hidden fund flow, which can at least partially solve the problems in the prior art.

In one aspect, the present invention provides a method for identifying hidden fund flow, including: according to first transaction data and full-amount customer transaction data after a target loan customer loans, the target loan customer is used as an initial node, a fund flow map of the target loan customer is constructed, customers participating in fund transaction are used as nodes in the fund flow map of the target loan customer, and fund transaction relations among the nodes are used as edges connecting the nodes; and identifying the recessive fund backflow information of the target loan customer according to the established fund flow map of the target loan customer, wherein the recessive fund backflow information is used for indicating whether the target loan customer has recessive fund backflow.

Optionally, the constructing a fund flow map of the target loan customer by using the target loan customer as an initial node according to the first transaction data after the target loan customer loans and the full-volume customer transaction data includes: according to the first transaction data after the target loan customer loans and the full-amount customer transaction data, taking the target loan customer as an initial node, and constructing an M-degree relationship fund flow map flowing from the target loan customer to other customers, wherein M is a positive integer; and constructing an N-degree relation fund flow map flowing from other customers to the target loan customer by taking the target loan customer as an initial node according to the full-amount customer transaction data, wherein N is a positive integer.

Optionally, the identifying the implicit fund flow information of the target loan customer according to the constructed fund flow map of the target loan customer includes: calculating the association degree between the client which is the leaf node in the N-degree relation fund flow map flowing from the target loan client to other clients and the client which is the leaf node in the M-degree relation fund flow map flowing from other clients to the target loan client according to the association degree calculation rule; and if the relevance exceeds a relevance threshold, determining that the target loan customer has hidden fund flow.

Optionally, the fund flow map flowing from the target loan customer to other customers is a 2-degree relationship fund flow map; the constructing of the M-degree relationship fund flow graph flowing from the target loan customer to other customers by taking the target loan customer as an initial node according to the first transaction data after the target loan customer loans and the full-volume customer transaction data comprises: according to the first transaction data after the target loan customer loans, the target loan customer is used as an initial node, a first customer who conducts first transaction with the target loan customer is used as a child node of the initial node, and the transaction relationship between the target loan customer and the first customer is used as a first degree edge, so that a fund flow map flowing from the target loan customer to the first customer is constructed; searching a second customer for trading with the first customer according to the full-amount customer trading data; and constructing a fund flow map flowing from the target loan customer to the second customer by taking the second customer as a leaf node of the child node and taking the transaction relationship between the first customer and the second customer as a second degree edge.

Optionally, the fund flow map flowing from other customers to the target loan customer is a 1-degree relationship fund flow map; according to the full-amount customer transaction data, with the target loan customer as a starting node, constructing an N-degree relationship fund flow map flowing from other customers to the target loan customer comprises the following steps: and according to the full-amount customer transaction data, taking the target loan customer as an initial node, taking a third customer which transfers funds to the target loan customer as a child node of the initial node, and taking the transaction relationship between the third customer and the target loan customer as a first degree, and constructing a fund flow map flowing from the third customer to the target loan customer.

Optionally, the searching for the second customer who transacts with the first customer according to the full-volume customer transaction data includes: and searching a second customer which is transacted with the first customer and meets the transaction time condition and the transaction amount condition in the full-volume customer transaction data.

Optionally, the method further includes: and when the target loan client is determined to have hidden fund flow, displaying an M-degree relation fund flow map from the target loan client to other clients and an N-degree relation fund flow map from other clients to the target loan client.

In another aspect, the present invention provides an apparatus for identifying hidden fund returns, including: the construction module is used for constructing a fund flow map of the target loan customer by taking the target loan customer as an initial node according to the first transaction data and the full amount customer transaction data after the target loan customer loans, wherein the fund flow map of the target loan customer takes each customer participating in fund transaction as a node and the fund transaction relationship among the nodes as an edge connecting the nodes; and the identification module is used for identifying the implicit fund reflux information of the target loan customer according to the established fund flow map of the target loan customer, wherein the implicit fund reflux information is used for indicating whether the target loan customer has implicit fund reflux or not.

Optionally, the building module is specifically configured to: according to the first transaction data after the target loan customer loans and the full-amount customer transaction data, taking the target loan customer as an initial node, and constructing an M-degree relationship fund flow map flowing from the target loan customer to other customers, wherein M is a positive integer; and constructing an N-degree relation fund flow map flowing from other customers to the target loan customer by taking the target loan customer as an initial node according to the full-amount customer transaction data, wherein N is a positive integer.

Optionally, the identification module is specifically configured to: calculating the association degree between the client which is the leaf node in the N-degree relation fund flow map flowing from the target loan client to other clients and the client which is the leaf node in the M-degree relation fund flow map flowing from other clients to the target loan client according to the association degree calculation rule; and if the relevance exceeds a relevance threshold, determining that the target loan customer has hidden fund flow.

Optionally, the fund flow map flowing from the target loan customer to other customers is a 2-degree relationship fund flow map; the construction module takes the target loan customer as an initial node according to the first transaction data after the target loan customer loans and the full-volume customer transaction data, and the construction of the M-degree relationship fund flow map flowing from the target loan customer to other customers comprises the following steps: according to the first transaction data after the target loan customer loans, the target loan customer is used as an initial node, a first customer who conducts first transaction with the target loan customer is used as a child node of the initial node, and the transaction relationship between the target loan customer and the first customer is used as a first degree edge, so that a fund flow map flowing from the target loan customer to the first customer is constructed; searching a second customer for trading with the first customer according to the full-amount customer trading data; and constructing a fund flow map flowing from the target loan customer to the second customer by taking the second customer as a leaf node of the child node and taking the transaction relationship between the first customer and the second customer as a second degree edge.

Optionally, the fund flow map flowing from other customers to the target loan customer is a 1-degree relationship fund flow map; the construction module takes the target loan client as an initial node according to the full-amount client transaction data, and the construction of the N-degree relationship fund flow map flowing from other clients to the target loan client comprises the following steps: and according to the full-amount customer transaction data, taking the target loan customer as an initial node, taking a third customer which transfers funds to the target loan customer as a child node of the initial node, and taking the transaction relationship between the third customer and the target loan customer as a first degree, and constructing a fund flow map flowing from the third customer to the target loan customer.

Optionally, the searching, by the building module, a second customer who transacts with the first customer according to the full-volume customer transaction data includes: and searching a second customer which is transacted with the first customer and meets the transaction time condition and the transaction amount condition in the full-volume customer transaction data.

Optionally, the apparatus further comprises: and the display module is used for displaying the M-degree relation fund flow map flowing from the target loan customer to other customers and the N-degree relation fund flow map flowing from other customers to the target loan customer when the target loan customer is determined to have hidden fund flow.

In another aspect, the present invention provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor executes the computer program to implement the steps of the method for identifying an implicit fund flow described in any of the above embodiments.

In yet another aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the steps of the method for identifying a hidden fund flow according to any one of the above embodiments.

The method and the device for identifying the recessive fund flow back, provided by the embodiment of the invention, can use a target loan customer as an initial node to construct the fund flow map of the target loan customer, and the fund flow map is constructed by taking the target loan customer as a core, so that the fund flow map can reflect the fund flow condition of the target loan customer at least to a certain extent, and whether the target loan customer has the recessive fund flow back can be identified according to the fund flow map, so that the behavior that the target loan customer uses loans in a violation way and the behavior of the illegal loan flow back is identified, and the risk monitoring in the banking credit field is enhanced.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts. In the drawings:

fig. 1 is a schematic illustration of the implicit reflow of loan funds by the loan client a.

Fig. 2 is a flowchart illustrating a method for identifying an implicit fund flow according to an embodiment of the present invention.

Fig. 3 is an M-degree relationship fund flow graph from a target loan customer to other customers, constructed by a Forward lookup (Forward Hash Search) method according to an embodiment of the present invention.

Fig. 4 is a schematic diagram of a fund flow link from a targeted loan customer to a second customer as constructed by an embodiment of the invention.

Fig. 5 is a schematic diagram of a fund flow graph from a target loan customer to other customers constructed according to an embodiment of the invention.

Fig. 6 is a flow case of first dispersing and then concentrating in a loan fund flow process.

Fig. 7 is a fund flow graph from other customers to the target loan customer, which is constructed by a reverse lookup (Backward Hash Search) method according to an embodiment of the present invention.

Fig. 8 is a partial flow chart of a method for identifying hidden fund returns according to an embodiment of the present invention.

Fig. 9 is a diagram illustrating a merging of leaf node clients from forward lookups and leaf node clients from reverse lookups according to an embodiment of the present invention.

Fig. 10 is a schematic diagram of a customer relevance calculation method according to an embodiment of the present invention.

Fig. 11 is a schematic structural diagram of an apparatus for identifying an implicit fund flow according to an embodiment of the present invention.

Fig. 12 is a schematic structural diagram of an apparatus for identifying hidden fund returns according to another embodiment of the present invention.

Fig. 13 is a schematic physical structure diagram of an electronic device according to a fourteenth embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the embodiments of the present invention are further described in detail below with reference to the accompanying drawings. The exemplary embodiments and descriptions of the present invention are provided to explain the present invention, but not to limit the present invention. It should be noted that the embodiments and features of the embodiments in the present application may be arbitrarily combined with each other without conflict.

The execution subject of the method for identifying the hidden fund flow back provided by the embodiment of the invention comprises but is not limited to a computer.

Fig. 2 is a schematic flow chart of a method for identifying an implicit fund flow according to an embodiment of the present invention, and as shown in fig. 2, the method for identifying an implicit fund flow according to an embodiment of the present invention includes:

s101, according to first transaction data and full-amount customer transaction data after a target loan customer loans, constructing a fund flow map of the target loan customer by taking the target loan customer as an initial node, wherein each customer participating in fund transaction in the fund flow map of the target loan customer is taken as a node, and a fund transaction relation between each node is taken as an edge connecting the nodes;

in this step, after the target loan customer loans for a period of time (for example, after the target loan customer loans for one month), the first transaction data and the full amount customer transaction data after the target loan customer loans are acquired; the first transaction data after the target loan client (borrower) loans refers to the transaction data of the target loan client (borrower) transferring the loaned funds to a transaction opponent by a certain reasonable reason; the full-amount customer transaction data may be transaction data of all customers in the local bank after the target loan customer loans, and the first transaction data after the target loan customer loans may be searched in the full-amount customer transaction data.

Since the fund flow of the target loan customer may involve multiple layers of transactions (as shown in fig. 1), the search for other transaction data related to the loan fund flow of the target loan customer may continue in the full amount customer transaction data after the first transaction data after the target loan customer's loan is obtained. Since each transaction relates to "borrower" and "lender", when constructing the fund flow graph of the target loan clients, the found clients participating in fund transactions in the transaction data can be used as nodes in the fund flow graph, and the fund transaction relationship between the nodes can be used as edges connecting the nodes.

S102, identifying implicit fund backflow information of the target loan customer according to the established fund flow map of the target loan customer, wherein the implicit fund backflow information is used for indicating whether the target loan customer has implicit fund backflow.

In the step, the fund flow map is constructed by taking the target loan customer as a core, so the fund flow map can reflect the fund flow condition of the target loan customer at least to a certain extent, and the implicit fund flow information of the target loan customer can be identified according to the fund flow map. The implicit fund flow information of the targeted loan customer may include: the target loan client does not generate the recessive fund flow, and the target loan client generates the recessive fund flow.

The method for identifying the hidden fund backflow provided by the embodiment of the invention can use a target loan customer as an initial node to construct the fund flow map of the target loan customer, and the fund flow map is constructed by taking the target loan customer as a core, so that the fund flow map can reflect the fund flow condition of the target loan customer at least to a certain extent, and whether the target loan customer has the hidden fund backflow can be identified according to the fund flow map, so that the target loan customer is identified to further identify the behaviors of illegal use of loans and illegal loan backflow, and the risk monitoring in the banking credit field is enhanced.

Optionally, the constructing a fund flow map of the target loan customer by using the target loan customer as an initial node according to the first transaction data after the target loan customer loans and the full-volume customer transaction data includes:

according to the first transaction data after the target loan customer loans and the full-amount customer transaction data, taking the target loan customer as an initial node, and constructing an M-degree relationship fund flow map flowing from the target loan customer to other customers, wherein M is a positive integer;

and constructing an N-degree relation fund flow map flowing from other customers to the target loan customer by taking the target loan customer as an initial node according to the full-amount customer transaction data, wherein N is a positive integer.

In this embodiment, with the target loan client as a starting node, an M-degree relational fund flow map (see a → B → C in fig. 3) flowing from the target loan client to other clients is constructed by a Forward lookup (forwarded Hash Search) method, and an N-degree relational fund flow map (see D → a in fig. 3) flowing from other clients to the target loan client is constructed by a reverse lookup method, so that a fund flow map (see fig. 3) with the target loan client as a core is restored, and the fund flow map with the target loan client as a core can be an open-loop map because a part of funds related to the target loan client may flow between other clients.

To facilitate the lookup of transaction data in the full amount customer transaction data regarding the flow of funds of the target loan customer, the full amount customer transaction data may be processed into a linked list, where each transaction data in the linked list should have a borrower customer, a transaction timestamp, a transaction amount, and a lender customer. For each transaction datum, a timestamp field for the next transaction for the same debit and credit may also be added, and nulled if no next transaction exists. In this way, it is convenient to find the time difference between the last two transactions of the same borrower and lender.

After the zipper table processing is completed, the SQL query method can be used for querying transaction data meeting transaction conditions in the zipper table.

Optionally, the fund flow map flowing from the target loan customer to other customers is a 2-degree relationship fund flow map; the constructing of the M-degree relationship fund flow graph flowing from the target loan customer to other customers by taking the target loan customer as an initial node according to the first transaction data after the target loan customer loans and the full-volume customer transaction data comprises:

according to the first transaction data after the target loan customer loans, the target loan customer is used as an initial node, a first customer who conducts first transaction with the target loan customer is used as a child node of the initial node, and the transaction relationship between the target loan customer and the first customer is used as a first degree edge, so that a fund flow map flowing from the target loan customer to the first customer is constructed;

searching a second customer for trading with the first customer according to the full-amount customer trading data;

and constructing a fund flow map flowing from the target loan customer to the second customer by taking the second customer as a leaf node of the child node and taking the transaction relationship between the first customer and the second customer as a second degree edge.

In this embodiment, the first transaction refers to a transaction in which the target loan client (borrower) transfers the funds credited to the target client, i.e., the first client, for some reasonable reason. The number of the first client is generally 1, the target loan client transfers the loan to the target client generally through one transaction, and a constructed fund flow map flowing from the target loan client to the first client is shown as A → B in FIG. 3, wherein A represents the target loan client, and B represents the first client.

Of course, in some special scenarios, the target loan client may transfer the loan funds to a plurality of target clients through a plurality of transactions, which is not limited in this embodiment.

When searching for a second customer who transacts with the first customer according to the full-volume customer transaction data, searching for a second customer who transacts with the first customer after the first customer transacts with the target loan customer in the full-volume customer transaction data by using transaction time as a searching condition, wherein the number of the second customers is at least 1. That is, the transaction time of the first customer and the second customer is after the first transaction between the target loan customer and the first customer.

There may be more than one fund transaction between the first customer and the second customer, and there may be duplication of fund flow links L (a, B, C) from the target loan customer to the second customer, constructed in transaction relationship (see fig. 4, where a represents the target loan customer, B represents the first customer, and C represents the target loan customer)₁、C₂Representing the second customer), in order to reduce unnecessary repeated searches and prevent exponential growth of leaf nodes, only the edge E (B, C) of the current layer may be focused and the edge E (B, C) may be deduplicated, finally resulting in the fund flow graph as shown in fig. 5.

Because in the 'hidden' fund backflow process, the following characteristics are provided: the amount transferred in each link is 80-120% of the loan amount of the loan client. Therefore, the fund transaction amount can be used as a search condition for searching each layer of nodes, namely when the fund transaction amount between the borrower client and the lender client is between 80% and 120% of the loan amount of the target loan client, the lender client is used as one of the nodes in the fund flow map of the target loan client.

At this time, the searching for the second customer who transacts with the first customer according to the full-volume customer transaction data may include: and searching a second customer which is transacted with the first customer and meets the transaction time condition and the transaction amount condition in the full-volume customer transaction data. In this embodiment, the transaction time condition may be after an initial transaction between the target loan client and the first client occurs; the transaction amount condition may be that the transaction amount is 80-120% of the loan amount of the target loan customer.

For the case of fund scatter and centralized circulation, as shown in the case of fig. 6, customer a gives customer B4 yuan money, customer B gives customer C4 yuan money, customer C transfers 4 yuan money to customer D through 3 transactions in the amount of 1 yuan and 2 yuan respectively, and finally customer D returns to customer a in a centralized manner. It is clear that this link does not comply with the monetary limit of the single link funds movement map. In FIG. 6, each node is followed by a doublet, e.g., "(2, 4)". The first element of the binary represents the current amount of money to be collected and the second element represents the cumulative amount of money to be collected from the previous customer. To deal with the situation, when the stepwise search is performed, each layer of search is performed by partitioning the edge data according to a starting point (debit cino) and an end point (credit cino), then sorting according to a timestamp (timestamp), and calculating and storing the accumulated amount (amount) of the current time. Finally, whether the accumulated amount is 80% -120% of the loan amount of the target loan client is also taken as one of the searching conditions.

And if the constructed fund flow map flowing from the target loan customer to other customers is a fund flow map with a relation of more than 2 degrees, after a fund flow link flowing from the target loan customer to the second customer is obtained, continuously searching a next-layer node D meeting the search condition to form E (C, D), and repeatedly searching the next layer.

Optionally, the fund flow map flowing from other customers to the target loan customer is a 1-degree relationship fund flow map; according to the full-amount customer transaction data, with the target loan customer as a starting node, constructing an N-degree relationship fund flow map flowing from other customers to the target loan customer comprises the following steps:

and according to the full-amount customer transaction data, taking the target loan customer as an initial node, taking a third customer which transfers funds to the target loan customer as a child node of the initial node, and taking the transaction relationship between the third customer and the target loan customer as a first degree, and constructing a fund flow map flowing from the third customer to the target loan customer.

In this embodiment, referring to fig. 7, the concept of Backward lookup (Backward Hash Search) is similar to that of forward lookup, and only the next-level node needs to be searched Backward according to the lookup condition, which is not described herein again.

Optionally, the method may further include: and when the target loan client is determined to have hidden fund flow, displaying an M-degree relation fund flow map from the target loan client to other clients and an N-degree relation fund flow map from other clients to the target loan client.

In this embodiment, when hidden fund backflow occurs to a target loan customer, the flow condition of the loan fund of the target loan customer can be displayed in a graphical manner, so that a supervisor can master the flow condition of the loan fund of the target loan customer.

If the nodes are subjected to duplication removal processing when the fund flow map of the target loan customer is constructed, the finally obtained fund flow map of the target loan customer can be restored, namely edges lost when the butted leaf nodes restore to the two ends respectively are removed from the two ends, and the full view of the fund flow condition of the target loan customer is restored.

As shown in fig. 8, optionally, the identifying the implicit fund flow information of the target loan customer according to the constructed fund flow map of the target loan customer includes:

s1021, calculating the association degree between the client which is a leaf node in the N-degree relation fund flow map flowing from the target loan client to other clients and the client which is a leaf node in the M-degree relation fund flow map flowing from other clients to the target loan client according to an association degree calculation rule;

in this step, as shown in fig. 9, the leaf node clients obtained by forward search and the leaf node clients obtained by reverse search are merged by using a relevancy calculation rule, where the relevancy calculation rule may be: judging whether the leaf node clients obtained by forward search and the leaf node clients obtained by reverse search are the same clients or not; or may be: and calculating the overlapping degree of the associated enterprise group of the leaf node client obtained by forward search and the associated enterprise group of the leaf node client obtained by reverse search, so as to judge whether an implicit transaction exists between the leaf node client obtained by forward search and the leaf node client obtained by reverse search, thereby carrying out butt joint.

Specifically, the Jaccard similarity may be used to determine the degree of association between the leaf node client obtained by the forward search and the leaf node client obtained by the reverse search. The calculation formula of the Jaccard coefficient is as follows:

wherein C represents a customer as a leaf node (leaf node customer found forward) in an N-degree relationship fund flow graph from the target loan customer to other customers;

d represents customers (leaf node customers obtained by reverse search) which are leaf nodes in the M-degree relationship fund flow graph from other customers to the target loan customer;

j (C, D) represents the degree of association of C and D.

Simply stated, it is the scale of the intersection of C and D divided by the scale of the union of C and D (see FIG. 10).

For example, we use the seven strong associations of client C as a set Γ and the seven strong associations of client D as a set Δ. And calculating the association degree of the two enterprises by dividing the length of the intersection of the gamma and the delta by the length of the union of the gamma and the delta, wherein the association degree of the client C and the client D is expressed as follows:

in the formula, L represents the length of the set, that is, the number of elements in the set.

And S1022, if the relevance exceeds a relevance threshold, determining that the target loan customer has hidden fund flow back.

In this step, when the association degree calculation rule is to determine whether the leaf node customer obtained by forward search and the leaf node customer obtained by reverse search are the same customer, and when the contact degree between the leaf node customer obtained by forward search and the leaf node customer obtained by reverse search is greater than the association degree threshold, it is determined that the target loan customer has hidden fund backflow.

When the association degree calculation rule is used for calculating the overlapping degree of the associated enterprise group of the leaf node client obtained by forward search and the associated enterprise group of the leaf node client obtained by reverse search, and the overlapping degree is greater than the association degree threshold value, the implicit transaction between the leaf node client obtained by forward search and the leaf node client obtained by reverse search can be determined, and therefore the target loan client is determined to have implicit fund return.

The method for identifying the hidden fund reflux provided by the invention supports the identification of the condition of dispersed and centralized circulation among customers; and the method can be realized by using a spark, flink and other distributed computing frameworks, and the query time control on hundred million-level data volume is completed within a minute level.

The two-way searching method provided by the invention can effectively identify the recessive fund reflux case by combining the algorithm of the similarity of the Jaccard, and the application field is wide. The method has the advantages of at least the following two aspects:

(1) it is possible to find transactions that are out-of-line with long links. According to the test, a five-degree long-line link backflow case can be found by adopting two-way search, and a hidden backflow case can be effectively identified by adopting a Jaccard similarity calculation method.

(2) The streaming case in a decentralized reconcentration can be dealt with. The method provided by the invention can also be completely identified for the condition that some customers rotate the loan score pen out and then reflow.

Fig. 11 is a schematic structural diagram of an apparatus for identifying hidden fund flow according to an embodiment of the present invention, and as shown in fig. 11, the apparatus for identifying hidden fund flow according to the embodiment of the present invention includes: the construction module 21 is configured to construct a fund flow map of the target loan customer by using the target loan customer as an initial node according to the first transaction data and the full-volume customer transaction data after the target loan customer loans, where in the fund flow map of the target loan customer, each customer participating in fund transaction is used as a node, and a fund transaction relationship between each node is used as an edge connecting the nodes; and the identification module 22 is used for identifying the implicit fund flow information of the target loan customer according to the established fund flow map of the target loan customer, wherein the implicit fund flow information is used for indicating whether the target loan customer has implicit fund flow.

The device for identifying the hidden fund backflow provided by the embodiment of the invention can use a target loan customer as an initial node to construct the fund flow map of the target loan customer, and the fund flow map is constructed by taking the target loan customer as a core, so that the fund flow map can reflect the fund flow condition of the target loan customer at least to a certain extent, and whether the target loan customer has the hidden fund backflow can be identified according to the fund flow map, so that the behavior that the target loan customer uses loans in violation and the behavior of the loan violation backflow is identified, and the risk monitoring in the banking credit field is enhanced.

As shown in fig. 12, optionally, the apparatus further includes: and the display module 23 is used for displaying the M-degree relation fund flow map flowing from the target loan customer to other customers and the N-degree relation fund flow map flowing from other customers to the target loan customer when the target loan customer is determined to have hidden fund flow.

The embodiment of the apparatus provided in the embodiment of the present invention may be specifically configured to execute the processing flows of the above method embodiments, and the functions of the apparatus are not described herein again, and refer to the detailed description of the above method embodiments.

It should be noted that the method and the apparatus for identifying the hidden fund flow back provided by the embodiment of the present invention may be used in the financial field, and may also be used in any technical field other than the financial field.

Fig. 13 is a schematic physical structure diagram of an electronic device according to an embodiment of the present invention, and as shown in fig. 13, the electronic device may include: a processor (processor)301, a communication Interface (communication Interface)302, a memory (memory)303 and a communication bus 304, wherein the processor 301, the communication Interface 302 and the memory 303 complete communication with each other through the communication bus 304. Processor 301 may call logic instructions in memory 303 to perform the following method: according to first transaction data and full-amount customer transaction data after a target loan customer loans, the target loan customer is used as an initial node, a fund flow map of the target loan customer is constructed, customers participating in fund transaction are used as nodes in the fund flow map of the target loan customer, and fund transaction relations among the nodes are used as edges connecting the nodes; and identifying the recessive fund backflow information of the target loan customer according to the established fund flow map of the target loan customer, wherein the recessive fund backflow information is used for indicating whether the target loan customer has recessive fund backflow.

In addition, the logic instructions in the memory 303 may be implemented in the form of software functional units and stored in a computer readable storage medium when the logic instructions are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.

The present embodiment discloses a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, enable the computer to perform the method provided by the above-mentioned method embodiments, for example, comprising: according to first transaction data and full-amount customer transaction data after a target loan customer loans, the target loan customer is used as an initial node, a fund flow map of the target loan customer is constructed, customers participating in fund transaction are used as nodes in the fund flow map of the target loan customer, and fund transaction relations among the nodes are used as edges connecting the nodes; and identifying the recessive fund backflow information of the target loan customer according to the established fund flow map of the target loan customer, wherein the recessive fund backflow information is used for indicating whether the target loan customer has recessive fund backflow.

The present embodiment provides a computer-readable storage medium, which stores a computer program, where the computer program causes the computer to execute the method provided by the above method embodiments, for example, the method includes: according to first transaction data and full-amount customer transaction data after a target loan customer loans, the target loan customer is used as an initial node, a fund flow map of the target loan customer is constructed, customers participating in fund transaction are used as nodes in the fund flow map of the target loan customer, and fund transaction relations among the nodes are used as edges connecting the nodes; and identifying the recessive fund backflow information of the target loan customer according to the established fund flow map of the target loan customer, wherein the recessive fund backflow information is used for indicating whether the target loan customer has recessive fund backflow.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In the description herein, reference to the description of the terms "one embodiment," "a particular embodiment," "some embodiments," "for example," "an example," "a particular example," or "some examples," etc., means 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.

The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.

Claims

1. A method of identifying hidden fund returns, comprising:

according to first transaction data and full-amount customer transaction data after a target loan customer loans, the target loan customer is used as an initial node, a fund flow map of the target loan customer is constructed, customers participating in fund transaction are used as nodes in the fund flow map of the target loan customer, and fund transaction relations among the nodes are used as edges connecting the nodes;

and identifying the recessive fund backflow information of the target loan customer according to the established fund flow map of the target loan customer, wherein the recessive fund backflow information is used for indicating whether the target loan customer has recessive fund backflow.

2. The method according to claim 1, wherein the constructing the fund flow map of the target loan customer according to the first transaction data after the target loan customer loan and the full amount customer transaction data by taking the target loan customer as a starting node comprises:

3. The method of claim 2, wherein the fund flow graph from other customers to the target loan customer is a 1 degree relationship fund flow graph;

according to the full-amount customer transaction data, with the target loan customer as a starting node, constructing an N-degree relationship fund flow map flowing from other customers to the target loan customer comprises the following steps:

4. The method according to claim 2, wherein the identifying the implicit fund flow information of the target loan customer according to the constructed fund flow map of the target loan customer comprises:

calculating the association degree between the client which is the leaf node in the N-degree relation fund flow map flowing from the target loan client to other clients and the client which is the leaf node in the M-degree relation fund flow map flowing from other clients to the target loan client according to the association degree calculation rule;

and if the relevance exceeds a relevance threshold, determining that the target loan customer has hidden fund flow.

5. The method of claim 4, further comprising:

and when the target loan client is determined to have hidden fund flow, displaying an M-degree relation fund flow map from the target loan client to other clients and an N-degree relation fund flow map from other clients to the target loan client.

6. The method of claim 2, wherein the fund flow graph from the target loan customer to other customers is a 2 degree relationship fund flow graph;

the constructing of the M-degree relationship fund flow graph flowing from the target loan customer to other customers by taking the target loan customer as an initial node according to the first transaction data after the target loan customer loans and the full-volume customer transaction data comprises:

7. The method of claim 6, wherein said locating a second customer that transacts with the first customer based on the full-size customer transaction data comprises:

and searching a second customer which is transacted with the first customer and meets the transaction time condition and the transaction amount condition in the full-volume customer transaction data.

8. An apparatus for identifying covert funds flow, comprising:

the construction module is used for constructing a fund flow map of the target loan customer by taking the target loan customer as an initial node according to the first transaction data and the full amount customer transaction data after the target loan customer loans, wherein the fund flow map of the target loan customer takes each customer participating in fund transaction as a node and the fund transaction relationship among the nodes as an edge connecting the nodes;

and the identification module is used for identifying the implicit fund reflux information of the target loan customer according to the established fund flow map of the target loan customer, wherein the implicit fund reflux information is used for indicating whether the target loan customer has implicit fund reflux or not.

9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method of any of claims 1 to 7 are implemented when the computer program is executed by the processor.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.