A method and device for mining corporate customers

Abstract

The present invention provides a method and device for mining corporate customers, which relates to the field of big data technology and can be used in the financial field or other technical fields. The method includes: establishing a capital flow diagram for corporate customers based on the transaction flows of each corporate customer in a preset time period; wherein the capital flow diagram includes each node and a directed edge connecting two nodes, the node corresponds to a corporate customer one-to-one, and each transaction flow corresponds to a directed edge; based on each node and the edge between each node, the mining parameters of corporate customers are obtained; according to the mining parameters of corporate customers and the corresponding corporate customer mining model, the mining reference value of each node is obtained; wherein the corporate customer mining model is preset; according to the mining reference value of each node and the corresponding mining threshold, the target corporate customer is obtained. The device is used to execute the above method. The method and device for mining corporate customers provided by the embodiment of the present invention save computing resources.

Description

Technical Field

[0001] The present invention relates to the field of big data technology, and in particular to a method and device for mining corporate customers. Background Art

[0002] For commercial banks, potential customer mining is a very important task. By mining potential customers, banking business can be expanded to improve bank efficiency.

[0003] Traditional customer mining mainly relies on account managers, which often has the problems of high cost and low efficiency. In the prior art, there are solutions for potential customer mining through big data training models, which require the collection of a large amount of customer information. On the one hand, some information is difficult to obtain, and on the other hand, model training requires data annotation, which takes a lot of time. Therefore, how to propose a method to more efficiently determine the potential corporate customers of bank products has become an important problem that needs to be solved in this field. Summary of the invention

[0004] In view of the problems in the prior art, an embodiment of the present invention provides a method and device for mining corporate customers, which can at least partially solve the problems in the prior art.

[0005] On the one hand, the present invention provides a method for mining corporate customers, comprising:

[0006] Based on the transaction flow of each corporate customer in a preset time period, a capital flow diagram of corporate customers is established; wherein the capital flow diagram includes each node and a directed edge connecting two nodes, the nodes correspond to corporate customers one by one, and each transaction flow corresponds to a directed edge;

[0007] Based on each node and the edges between nodes, obtain the mining parameters for corporate customers;

[0008] According to the mining parameters of corporate customers and the corresponding corporate customer mining model, a mining reference value of each node is obtained; wherein the corporate customer mining model is preset;

[0009] According to the mining reference value of each node and the corresponding mining threshold, the target corporate customers are obtained.

[0010] In another aspect, the present invention provides an excavation device for corporate customers, comprising:

[0011] Establishing a module for establishing a capital flow diagram for corporate customers based on the transaction flow of each corporate customer in a preset time period; wherein the capital flow diagram includes each node and a directed edge connecting two nodes, the nodes correspond to corporate customers one by one, and each transaction flow corresponds to a directed edge;

[0012] A first acquisition module is used to obtain mining parameters of corporate customers based on each node and the edges between each node;

[0013] A second obtaining module is used to obtain a mining reference value of each node according to mining parameters of corporate customers and a corresponding corporate customer mining model; wherein the corporate customer mining model is preset;

[0014] The third acquisition module is used to obtain target corporate customers according to the mining reference value of each node and the corresponding mining threshold.

[0015] On the other hand, the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the corporate customer mining method described in any of the above embodiments when executing the program.

[0016] On the other hand, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for mining corporate customers described in any of the above embodiments.

[0017] The method and device for mining corporate customers provided by the embodiments of the present invention can establish a capital flow diagram for corporate customers based on the transaction flows of each corporate customer in a preset time period, obtain mining parameters for corporate customers based on each node and the edges between each node, obtain a mining reference value for each node based on the mining parameters for corporate customers and a corresponding corporate customer mining model, obtain a target corporate customer based on the mining reference value of each node and a corresponding mining threshold, and mine the target corporate customer from a large number of corporate customers. Since there is no need for model training, the demand for hardware can be reduced, saving computing resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0019] Figure 1 It is a flowchart of the corporate customer mining method provided by the first embodiment of the present invention.

[0020] Figure 2 It is a partial schematic diagram of the fund flow diagram for corporate customers provided by the second embodiment of the present invention.

[0021] Figure 3It is a flowchart of a method for mining corporate customers provided in the third embodiment of the present invention.

[0022] Figure 4 It is a schematic diagram of the structure of an excavation device for corporate customers provided by the fourth embodiment of the present invention.

[0023] Figure 5 It is a schematic diagram of the structure of an excavation device for corporate customers provided by the fifth embodiment of the present invention.

[0024] Figure 6 It is a schematic diagram of the structure of an excavation device for corporate customers provided by the sixth embodiment of the present invention.

[0025] Figure 7 It is a schematic diagram of the structure of an excavation device for corporate customers provided by the seventh embodiment of the present invention.

[0026] Figure 8 It is a schematic diagram of the physical structure of an electronic device provided by the eighth embodiment of the present invention. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical scheme and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention. It should be noted that, in the absence of conflict, the embodiments in this application and the features in the embodiments can be combined with each other arbitrarily.

[0028] The following takes a server as an example of an execution subject to illustrate the implementation process of the method for mining corporate customers provided by an embodiment of the present invention.

[0029] Figure 1 FIG. 1 is a flow chart of a method for mining corporate customers provided by the first embodiment of the present invention. Figure 1 As shown, the method for mining corporate customers provided by the embodiment of the present invention includes:

[0030] S101. Based on the transaction flow of each corporate customer in a preset time period, a capital flow graph of corporate customers is established; wherein the capital flow graph includes each node and a directed edge connecting two nodes, the nodes correspond to corporate customers one by one, and each transaction flow corresponds to a directed edge;

[0031] Specifically, the server can establish a fund flow diagram for corporate customers based on the transaction flows of each corporate customer in a preset time period, with the two parties in the transaction flow as nodes, drawing edges connecting the two nodes, and with the fund flow as the direction of the edge. The fund flow diagram includes each node and a directed edge connecting the two nodes. Among them, the nodes correspond one-to-one to corporate customers, each transaction flow corresponds to a directed edge, the direction of the directed edge is from the fund outflow party to the fund inflow party, and the transaction amount of each transaction flow is used as the transaction amount of the corresponding directed edge. The preset time period is set according to actual needs, such as being set to 1 year, and the embodiment of the present invention does not limit it.

[0032] It should be noted that the transaction partner that has transaction flow with the corporate customer may not be a customer of the bank where the corporate customer is located, so there are nodes in the corporate customer's fund flow diagram that correspond to corporate customers that are not the bank where the corporate customer is located.

[0033] For example, Figure 2 FIG. 1 is a partial schematic diagram of a fund flow diagram for corporate customers provided by the second embodiment of the present invention. Figure 2 As shown in the figure, there is a transaction flow between node A and node B, directed edge 1 indicates that funds flow from node A to node B, directed edges 5 and 6 indicate that funds flow from node B to node A; there is a transaction flow between node B and node C, directed edges 2 and 7 indicate that funds flow from node B to node C; there is a transaction flow between node A and node C, directed edge 4 indicates that funds flow from node A to node C, directed edges 3 and 8 indicate that funds flow from node C to node A. Each of the above directed edges has a corresponding transaction amount.

[0034] S102, obtaining mining parameters for corporate customers based on each node and the edges between each node;

[0035] Specifically, the server can obtain mining parameters for each node according to each node and the edges between each node in the capital flow graph of corporate customers. The mining parameters are set according to actual needs and are not limited in the embodiment of the present invention.

[0036] For example, the mining parameters for corporate customers may include the out-degree and in-degree of each node, the frequency of capital inflow and the frequency of capital outflow of each node, and the total amount of capital outflow and the total amount of capital inflow of each node.

[0037] For example, the mining parameters for corporate customers may include the degree of each node, the total transaction amount corresponding to each node, and the transaction frequency corresponding to each node.

[0038] S103, obtaining a mining reference value of each node according to the mining parameters of corporate customers and the corresponding corporate customer mining model; wherein the corporate customer mining model is preset;

[0039] Specifically, the server inputs the mining parameters of corporate customers into the corresponding corporate customer mining model, and can output the mining reference value of each node. The corporate customer mining model is preset.

[0040] S104: Obtain target corporate customers according to the mining reference value of each node and the corresponding mining threshold.

[0041] Specifically, the server can obtain target corporate customers according to the mining reference values ​​of each node and the corresponding mining threshold. The obtained target corporate customers can be regarded as key corporate customers or high-value customers, and financial products can be recommended to them. The mining threshold is set according to actual needs and is not limited in the embodiment of the present invention.

[0042] For example, the server compares the mining reference value of each node with the corresponding mining threshold, and takes the corporate customers corresponding to the nodes greater than the mining threshold as the target corporate customers.

[0043] For example, the mining threshold is 10%, and the server sorts the mining reference values ​​of each node in descending order, and takes the corporate customers corresponding to each node in the top 10% of the sort as the target corporate customers.

[0044] The method for mining corporate customers provided by the embodiment of the present invention can establish a capital flow diagram for corporate customers based on the transaction flow of each corporate customer in a preset time period, obtain the mining parameters for corporate customers based on each node and the edges between each node, obtain the mining reference value of each node based on the mining parameters for corporate customers and the corresponding corporate customer mining model, obtain the target corporate customer based on the mining reference value of each node and the corresponding mining threshold, and mine the target corporate customer from a large number of corporate customers. Since there is no need to train the model, the demand for hardware can be reduced, saving computing resources. In addition, the mining efficiency of corporate customers is also improved. On the basis of the above embodiments, further, the mining parameters for corporate customers include the out-degree and in-degree of each node, the frequency of capital inflow and outflow of each node, and the total amount of outflow and inflow of each node; accordingly, the mining parameters for corporate customers obtained based on the edges between each node include:

[0045] The number of different nodes connected to each node through the capital outflow edge is counted as the out-degree of each node, and the number of different nodes connected to each node through the capital inflow edge is counted as the in-degree of each node;

[0046] Calculate the quotient of the number of capital inflow edges of each node and the preset time period as the capital inflow frequency of each node, and calculate the quotient of the number of capital outflow edges of each node and the preset time period as the capital outflow frequency of each node;

[0047] Calculate the sum of the transaction amounts corresponding to each capital inflow edge of each node as the total capital inflow of each node, and calculate the sum of the transaction amounts corresponding to each capital outflow edge of each node as the total capital outflow of each node.

[0048] Specifically, the mining parameters for corporate customers include the out-degree and in-degree of each node, the frequency of capital inflow and the frequency of capital outflow of each node, and the total amount of capital outflow and the total amount of capital inflow of each node.

[0049] The server may count the number of different nodes connected to each node through the outflow edge of funds in the capital flow graph of the corporate customer as the out-degree of each node, and count the number of different nodes connected to each node through the inflow edge of funds as the in-degree of each node. The outflow edge of a node refers to a directed edge where funds flow out of the node, and the inflow edge of a node refers to a directed edge where funds flow into the node.

[0050] The server may obtain the number of fund inflow edges of each node from the fund flow graph of the corporate customer, and then calculate the quotient of the number of fund inflow edges of each node and the preset time period as the fund inflow frequency of each node. The server may obtain the number of fund outflow edges of each node from the fund flow graph of the corporate customer, and then calculate the quotient of the number of fund outflow edges of each node and the preset time period as the fund outflow frequency of each node.

[0051] The server may obtain the transaction amount corresponding to each capital inflow edge of each node in the capital flow diagram of the corporate customer, and then calculate the sum of the transaction amounts corresponding to each capital inflow edge of each node as the total capital inflow of each node. The server may obtain the transaction amount corresponding to each capital outflow edge of each node in the capital flow diagram of the corporate customer, and then calculate the sum of the transaction amounts corresponding to each capital outflow edge of each node as the total capital outflow of each node.

[0052] For example, in Figure 2In the example, node A has four capital inflow edges: directed edge 3, directed edge 5, directed edge 6, and directed edge 8, and node A has two capital outflow edges: directed edge 1 and directed edge 4. The different nodes connected to node A through the capital outflow edges are node B and node C, so the in-degree of node A can be obtained as 2; the different nodes connected to node A through the capital inflow edges are node B and node C, so the out-degree of node A is 2. The preset time period is t, then the capital inflow frequency of node A is 4 / t, and the capital outflow frequency of node A is 2 / t. The transaction amount corresponding to directed edge 3 is obtained as Y3, the transaction amount corresponding to directed edge 5 is obtained as Y5, the transaction amount corresponding to directed edge 6 is obtained as Y6, and the transaction amount corresponding to directed edge 8 is obtained as Y8, then the total capital inflow of node A is Q1=Y3+Y5+Y6+Y8. The transaction amount corresponding to directed edge 1 is Y1, and the transaction amount corresponding to directed edge 4 is Y4. Then the total capital outflow of node A is Q2=Y1+Y4.

[0053] On the basis of the above embodiments, further, the corporate customer mining model is a core enterprise model, and the core enterprise model includes:

[0054] I(i)=In_I(i)×Out_I(i) (1)

[0055]

[0056]

[0057] Where I(i) represents the mining reference value, In_Degree(i) represents the in-degree of the i-th node, In_Freq(i) represents the frequency of capital inflow of the i-th node, In_Amt(i) represents the total amount of capital inflow of the i-th node, Out_Degree(i) represents the out-degree of the i-th node, Out_Freq(i) represents the frequency of capital outflow of the i-th node, Out_Amt(i) represents the total amount of capital outflow of the i-th node, In_Freq(i) represents the frequency of capital outflow of the i-th node, Out_Amt(i) represents the total amount of capital outflow of the i-th node, max Indicates the maximum value of the node's capital inflow frequency, In_Freq min Indicates the minimum frequency of capital inflow into the node, In_Amt max Indicates the maximum amount of capital inflow into the node, In_Amt min Indicates the minimum value of the total amount of funds flowing into the node, Out_Freq max Indicates the maximum value of the node's fund outflow frequency, Out_Freq min Indicates the minimum value of the node's fund outflow frequency, Out_Amt max Indicates the maximum amount of funds outflow of the node, Out_Amt minRepresents the minimum value of the total amount of capital outflow of the node, N represents the total number of nodes, i is a positive integer and i is less than or equal to N.

[0058] Specifically, in formula (1), In_I(i) represents the in-degree influence of the i-th node, Out_I(i) represents the out-degree influence of the i-th node, and I(i) is used to measure the influence of the enterprise customer corresponding to the i-th node. It is used to measure the influence of the frequency of capital inflow into the node. In formula (2) It is used to measure the influence of the total amount of capital inflow of the node. In formula (3) It is used to measure the influence of the frequency of capital outflow of the node. In formula (3) Used to measure the influence of the total amount of funds outflow of the node.

[0059] The server compares the fund inflow frequency of each node and obtains the minimum fund inflow frequency as In_Freq min , and obtain the maximum fund inflow frequency as In_Freq max The server compares the fund outflow frequency of each node and obtains the minimum fund outflow frequency as Out_Freq min , and obtain the maximum fund outflow frequency as Out_Freq max The server compares the total amount of funds inflow of each node and obtains the minimum total amount of funds inflow as In_Amt min , get the maximum total inflow as In_Amt max The server compares the total outflow of funds of each node and obtains the minimum total outflow of funds as Out_Amt min , get the maximum total inflow as Out_Amt max .

[0060] For the i-th node, the server calculates the in-degree In_Degree(i) of the i-th node, the frequency of capital inflow In_Freq(i) of the i-th node, the total amount of capital inflow In_Amt(i) of the i-th node, and In_Freq(i) of the i-th node. min 、In_Freq max 、In_Amt min and In_Amt max Substituting into formula (2), the in-degree influence In_I(i) of the i-th node can be calculated. The server calculates the out-degree Out_Degree(i) of the i-th node, the outflow frequency Out_Freq(i) of the i-th node, the total outflow amount Out_Amt(i) of the i-th node, and Out_Freq(i) of the i-th node. min、Out_Freq max 、Out_Amt min and Out_Amt max Substituting this into formula (3), we can calculate the out-degree influence Out_I(i) of the i-th node. Then, we can calculate the product of In_I(i) and Out_I(i) using formula (1).

[0061] The mining reference value I(i) of each node obtained through the core enterprise model can be used to judge the influence of the corporate customers corresponding to the node. The larger the mining reference value, the greater the influence of the enterprise corresponding to the node, and it will have the characteristics of a core enterprise in the industrial chain. By marketing to the target corporate customers obtained based on the core enterprise model, the small and medium-sized enterprises around the target corporate customers can be radiated to achieve point-to-point chain.

[0062] On the basis of the above embodiments, further, the mining parameters for corporate customers include the degree of each node, the total transaction amount corresponding to each node, and the transaction frequency corresponding to each node; wherein the node pair refers to two adjacent nodes; accordingly, the mining parameters for corporate customers obtained based on the edges between each node include:

[0063] Count the number of adjacent nodes of each node and obtain the degree of each node;

[0064] Calculate the sum of the transaction amounts corresponding to each directed edge between the two nodes in each node pair as the total transaction amount of each node pair;

[0065] The quotient of the number of directed edges between two nodes in each node pair and the preset time period is calculated as the transaction frequency of each node pair.

[0066] Specifically, the mining parameters of corporate customers include the degree of each node, the total transaction amount of each node pair, and the transaction frequency of each node pair. A node pair refers to two adjacent nodes, that is, two nodes directly connected by a directed edge. There are many node pairs in the capital flow graph of corporate customers.

[0067] The server can count the number of adjacent nodes of each node in the fund flow diagram of the corporate customer as the degree of each node. When counting adjacent nodes, the same node is counted only once.

[0068] The server can obtain the transaction amount corresponding to each directed edge between two nodes in each node pair in the capital flow diagram of the corporate customer, and then calculate the sum of the transaction amounts corresponding to each directed edge between two nodes in each node pair, and use the sum result as the total transaction amount of each node pair.

[0069] The server can obtain the number of directed edges between two adjacent nodes in each node pair from the capital flow diagram of the corporate customers, and then divide the number of directed edges obtained by the statistics by the preset time period, and use the calculated quotient as the capital outflow frequency of each node.

[0070] For example, in Figure 2 In the example, node A is adjacent to node B and node C, node B is adjacent to node C, node A and node B form node pair 1, node A and node C form node pair 2, and node B and node C form node pair 3. Node A has four capital inflow edges: directed edge 3, directed edge 5, directed edge 6, and directed edge 8, and node A has two capital outflow edges: directed edge 1 and directed edge 4. Therefore, the degree of node A is equal to 2. For node pair 1, there are three directed edges between node A and node B, namely directed edge 1, directed edge 5, and directed edge 6. The preset time period is t, so the transaction frequency of node pair 1 is 3 / t. The transaction amount corresponding to directed edge 1 is Y1, the transaction amount corresponding to directed edge 5 is Y5, and the transaction amount corresponding to directed edge 6 is Y6. Then the total transaction amount of node pair 1 is P1=Y1+Y5+Y6. Similarly, the transaction frequency and total transaction amount of node pair 2, as well as the transaction frequency and total transaction amount of node pair 3 can be calculated.

[0071] On the basis of the above embodiments, further, the corporate customer mining model is an enterprise influence model, and the enterprise influence model includes:

[0072]

[0073]

[0074] Among them, Freq(i,j) represents the transaction frequency of node pair (i,j), Amt(i,j) represents the total transaction amount of node pair (i,j), weight(i,j) represents the weight of node pair (i,j), Freq max Indicates the maximum transaction frequency of each node pair, Freq min Indicates the minimum transaction frequency of each node pair, Amt max Indicates the maximum transaction amount of each node pair, Amt min represents the minimum value of the total transaction amount of each node pair, m represents the number of adjacent nodes of the i-th node, degree(j) represents the degree of the j-th node adjacent to the i-th node, PR(i) represents the influence value of the i-th node, PR(j) represents the influence value of the j-th node adjacent to the i-th node, d is a constant, d is greater than 0 and less than 1, N represents the total number of nodes, i is a positive integer and i is less than or equal to N, j is a positive integer and j is less than or equal to m.

[0075] Specifically, the server compares the transaction frequencies of each node pair and obtains the minimum transaction frequency as Freq min , and obtain the maximum transaction frequency as Freq max The server compares the transaction amounts of each node pair and obtains the minimum transaction amount as Amt min , and get the maximum transaction frequency as Amt max .

[0076] For the node pair (i, j), the server sends the total transaction amount of the node pair (i, j), the transaction frequency Freq(i, j), and Freq min 、Freq max 、Amt min and Amt max Substituting this into formula (4), we can calculate the weight (i, j) of the node pair (i, j).

[0077] Formula (5) uses an improved PageRank algorithm. Formula (5) is an iterative formula. It is necessary to know the initial influence value of each node. The initial influence value of each node can be set to 1. d is set according to actual experience, and the embodiment of the present invention does not limit it. After obtaining the weights of each node pair, the first round of iterative calculation is performed through formula (5). The influence value of each node after the first round of iteration can be calculated, and then the difference between the influence value of each node after the first round of iteration and the initial influence value is calculated. The above difference is compared with the cutoff threshold. If the above difference is less than the cutoff threshold, the influence value of the node after the first round of iteration can be used as the mining reference value of the node. If the above difference is greater than or equal to the cutoff threshold, the next round of iterative calculation is performed. By performing the second round of iterative calculations using formula (5), the influence value of each node after the second round of iteration can be calculated, and then the difference between the influence value of each node after the second round of iteration and the influence value after the first round of iteration is calculated, and the above difference is compared with the cutoff threshold. If the above difference is less than the cutoff threshold, the influence value of the node after the second round of iteration can be used as the mining reference value of the node. If the above difference is greater than or equal to the cutoff threshold, the next round of iterative calculation is performed. And so on, until the mining reference values ​​of all nodes are obtained. Among them, the cutoff threshold is preset and set according to actual experience, and the embodiment of the present invention does not limit it.

[0078] After providing the degree of each node, the total transaction amount corresponding to each node, and the transaction frequency corresponding to each node to the enterprise influence model, the final output is the mining reference value of each node after processing by the enterprise influence model.

[0079] The mining reference value of each node obtained through the enterprise influence model can be used to judge the influence of the enterprise customers corresponding to the node. The larger the mining reference value, the greater the influence of the enterprise corresponding to the node.

[0080] Figure 3 FIG. 4 is a flow chart of a method for mining corporate customers provided by a third embodiment of the present invention. Figure 3 As shown, based on the above embodiments, the method for mining corporate customers provided by the embodiment of the present invention further includes:

[0081] S301, screening out external corporate customers and internal corporate customers from corporate customers corresponding to each node;

[0082] Specifically, the server can filter out external corporate customers and internal corporate customers from the corporate customers corresponding to each node in the capital flow diagram of corporate customers. The external corporate customers refer to corporate customers that are not local banks, that is, companies that have not opened corporate accounts in local banks; the internal corporate customers refer to corporate customers of local banks, that is, companies that have opened corporate accounts in local banks. Local banks refer to banks that provide transaction flows for corporate customers.

[0083] S302, obtaining the holding status of each internal corporate customer's local financial product;

[0084] Specifically, the server can query each internal corporate customer in the local banking system whether the internal corporate customer holds a local financial product. If the internal corporate customer holds a local financial product, the corresponding local financial product holding status is held, which can be represented by 1. If the internal corporate customer does not hold a local financial product, the corresponding local financial product holding status is held or not held, which can be represented by 0.

[0085] S303, obtaining the interest level value of the external corporate customer in the local financial product according to the holding status of the local financial product of the internal corporate customer corresponding to each node adjacent to the node corresponding to the external corporate customer and the weight of the node pair corresponding to the node corresponding to the external corporate customer;

[0086] Specifically, for any node corresponding to an external corporate customer selected, the server can obtain the node adjacent to the node corresponding to the external corporate customer from the capital flow diagram of the corporate customer, and obtain the node corresponding to the internal corporate customer from the node adjacent to the node corresponding to the external corporate customer, so as to obtain the internal corporate customers corresponding to each node adjacent to the node corresponding to the external corporate customer, and then according to the holding status of the local financial products of the internal corporate customers and each internal corporate customer, the holding status of the local financial products of the internal corporate customers can be obtained, that is, the holding status of the local financial products of the internal corporate customers corresponding to each node adjacent to the node corresponding to the external corporate customer is obtained. The server can obtain the interest level value of the external corporate customer in the local financial product according to the holding status of the local financial products of the internal corporate customers corresponding to each node adjacent to the node corresponding to the external corporate customer and the weight of the node pair corresponding to the node corresponding to the external corporate customer. Among them, the weight of the node pair corresponding to the node corresponding to the external corporate customer refers to the weight of each node pair consisting of the node corresponding to the external corporate customer and the node corresponding to the internal corporate customer.

[0087] For example, the node D corresponding to the external corporate customer d has 10 adjacent nodes in the corporate customer fund flow diagram, and 7 of the 10 adjacent nodes correspond to internal corporate customers. Then the server will obtain the holding status of the local financial products of the internal corporate customers corresponding to the above 7 nodes. Node D and the above 7 nodes can form 7 node pairs, and the weight of the node pair corresponding to the node D corresponding to the external corporate customer d is the weight of the above 7 node pairs, with a total of 7 weights.

[0088] S304: If it is determined that the interest level of the external corporate customer in local finance is greater than a recommendation threshold, recommend the local financial product to the external corporate customer.

[0089] Specifically, after obtaining the interest level value of the external corporate customer in local finance, the server compares the interest level value of the external corporate customer in local finance with the recommendation threshold. If the interest level value of the external corporate customer in local finance is greater than the recommendation threshold, the local financial product can be recommended to the external corporate customer by email or other means. It is understandable that if the interest level value of the external corporate customer in local finance is less than or equal to the recommendation threshold, the local financial product will not be recommended to the external corporate customer. Among them, the recommendation threshold is set according to actual needs and is not limited in the embodiment of the present invention.

[0090] Steps S301 to S304 can recommend local financial products held by internal corporate customers with whom the external corporate customers have a large number of transactions, which is conducive to discovering high-value customers from external corporate customers.

[0091] On the basis of the above embodiments, further, obtaining the interest level value of the external corporate customer in the local financial product according to the holding status of the local financial product of the corporate customer corresponding to each node adjacent to the node corresponding to the external corporate customer and the weight of the node pair corresponding to the node corresponding to the external corporate customer includes:

[0092] According to the formula Calculate and obtain the interest level value p(k,u) of external corporate customer k in local financial product u, where weight(k,j) represents the weight of the node pair (k,j), d(j,u) represents the holding status of the internal corporate customers corresponding to the j-th adjacent node of external corporate customer k in local financial product u, and m represents the number of internal corporate customers corresponding to the adjacent nodes of external corporate customer k.

[0093] Specifically, the server brings the holding status of the local financial product u of the internal corporate customer corresponding to the jth adjacent node of the external corporate customer k and the weight of the node pair (k, j) into the formula The interest level value p(k,u) of external corporate customer k in local financial product u can be calculated. The number of weights of node pairs (k,j) is m.

[0094] For example, the node K corresponding to the external corporate customer k has 6 adjacent nodes in the corporate customer funds flow diagram, and 5 of the 6 adjacent nodes correspond to internal corporate customers. Then the number of internal corporate customers corresponding to the adjacent nodes of the external corporate customer k is 5, that is, m=5. The server will obtain the holding status of the local financial product u of the internal corporate customers corresponding to the above 5 nodes, which are recorded as 1, 1, 1, 1 and 0 respectively. 0 means that the local financial product u is not held, and 1 means that the local financial product u is held. Node K and the above 5 nodes can form 5 node pairs. The weight of the node pair corresponding to the node K corresponding to the external corporate customer d is the weight of the above 5 node pairs. There are 5 weights in total, which are recorded as w1, w2, w3, w4 and w5 respectively. w1, w2, w3, w4 correspond to 1 respectively, and w5 corresponds to 0. The server substitutes w1, w2, w3, w4 and w5 and their corresponding holding status into the formula In the equation (1), we obtain p(k,u)=w1×1+w2×1+w3×1+w4×1+w0×1, and we can calculate the interest level p(k,u) of external corporate customer k in local financial product u.

[0095] Figure 4Schematic diagram of the structure of the excavation device for corporate customers provided by the fourth embodiment of the present invention. Figure 4 As shown, the corporate customer mining device provided by the embodiment of the present invention includes a building module 401, a first obtaining module 402, a second obtaining module 403 and a third obtaining module 404, wherein:

[0096] Establishing module 401 is used to establish a capital flow diagram for corporate customers based on the transaction flows of each corporate customer in a preset time period; wherein the capital flow diagram includes each node and a directed edge connecting two nodes, the node corresponds to a corporate customer one-to-one, and each transaction flow corresponds to a directed edge; the first obtaining module 402 is used to obtain mining parameters for corporate customers based on each node and the edges between each node; the second obtaining module 403 is used to obtain a mining reference value for each node based on the mining parameters for corporate customers and a corresponding corporate customer mining model; wherein the corporate customer mining model is preset; the third obtaining module 404 is used to obtain the target corporate customer based on the mining reference value of each node and a corresponding mining threshold.

[0097] Specifically, the establishment module 401 can establish a fund flow diagram for corporate customers based on the transaction flows of each corporate customer in a preset time period, with the transaction parties in the transaction flow as nodes, drawing edges connecting the two nodes, and taking the fund flow as the direction of the edge, wherein the fund flow diagram includes each node and a directed edge connecting the two nodes. Among them, the nodes correspond to corporate customers one by one, each transaction flow corresponds to a directed edge, the direction of the directed edge is from the fund outflow party to the fund inflow party, and the transaction amount of each transaction flow is used as the transaction amount of the corresponding directed edge. The preset time period is set according to actual needs, such as being set to 1 year, and the embodiment of the present invention does not limit it.

[0098] The first obtaining module 402 can obtain mining parameters for each node according to each node and the edges between each node in the capital flow diagram of corporate customers. The mining parameters are set according to actual needs and are not limited in the embodiment of the present invention.

[0099] The second obtaining module 403 inputs the mining parameters of corporate customers into the corresponding corporate customer mining model, and can output the mining reference value of each node. The corporate customer mining model is preset.

[0100] The third acquisition module 404 can obtain target corporate customers according to the mining reference value of each node and the corresponding mining threshold. The obtained target corporate customers can be regarded as key corporate customers or high-value customers, and financial products can be recommended to them. The mining threshold is set according to actual needs and is not limited in the embodiment of the present invention.

[0101] The mining device for corporate customers provided by the embodiment of the present invention can establish a capital flow diagram for corporate customers based on the transaction flow of each corporate customer in a preset time period, obtain mining parameters for corporate customers based on each node and the edge between each node, obtain mining reference values ​​for each node based on mining parameters for corporate customers and corresponding mining models for corporate customers, obtain target corporate customers based on mining reference values ​​for each node and corresponding mining thresholds, and mine target corporate customers from a large number of corporate customers. Since there is no need to train a model, the demand for hardware can be reduced, saving computing resources. In addition, the mining efficiency for corporate customers is also improved.

[0102] Figure 5 Schematic diagram of the structure of the excavation device for corporate customers provided by the fifth embodiment of the present invention. Figure 5 As shown, on the basis of the above embodiments, further, the mining parameters of the corporate customers include the out-degree and in-degree of each node, the fund inflow frequency and fund outflow frequency of each node, the total outflow fund amount and the total inflow fund amount of each node; accordingly, the first obtaining module 402 includes a statistical unit 4021, a first calculation unit 4022 and a second calculation unit 4023, wherein:

[0103] The statistical unit 4021 is used to count the number of different nodes connected to each node through the capital outflow edge as the out-degree of each node, and to count the number of different nodes connected to each node through the capital inflow edge as the in-degree of each node; the first calculation unit 4022 is used to calculate the quotient of the number of capital inflow edges of each node and the preset time period as the capital inflow frequency of each node, and to calculate the quotient of the number of capital outflow edges of each node and the preset time period as the capital outflow frequency of each node; the second calculation unit 4023 is used to calculate the sum of the transaction amounts corresponding to each capital inflow edge of each node as the total capital inflow amount of each node, and to calculate the sum of the transaction amounts corresponding to each capital outflow edge of each node as the total capital outflow amount of each node.

[0104] On the basis of the above embodiments, further, the corporate customer mining model is a core enterprise model, and the core enterprise model includes:

[0105] I(i)=In_I(i)×Out_I(i)

[0106]

[0107]

[0108] Where I(i) represents the mining reference value, In_Degree(i) represents the in-degree of the i-th node, In_Freq(i) represents the frequency of capital inflow of the i-th node, In_Amt(i) represents the total amount of capital inflow of the i-th node, Out_Degree(i) represents the out-degree of the i-th node, Out_Freq(i) represents the frequency of capital outflow of the i-th node, Out_Amt(i) represents the total amount of capital outflow of the i-th node, In_Freq(i) represents the frequency of capital outflow of the i-th node, Out_Amt(i) represents the total amount of capital outflow of the i-th node, max Indicates the maximum value of the node's capital inflow frequency, In_Freq min Indicates the minimum frequency of capital inflow into the node, In_Amt max Indicates the maximum amount of capital inflow into the node, In_Amt min Indicates the minimum value of the total amount of funds flowing into the node, Out_Freq max Indicates the maximum value of the node's fund outflow frequency, Out_Freq min Indicates the minimum value of the node's fund outflow frequency, Out_Amt max Indicates the maximum amount of funds outflow of the node, Out_Amt min Represents the minimum value of the total amount of capital outflow of the node, N represents the total number of nodes, i is a positive integer and i is less than or equal to N.

[0109] Figure 6 : is a schematic diagram of the structure of the excavation device for corporate customers provided by the sixth embodiment of the present invention, such as Figure 6 As shown, on the basis of the above embodiments, further, the mining parameters of the corporate customers include the degree of each node, the total transaction amount corresponding to each node, and the transaction frequency corresponding to each node; wherein the node pair refers to two adjacent nodes; accordingly, the first obtaining module 402 includes a third calculating unit 4024, a fourth calculating unit 4025 and a fifth calculating unit 4026, wherein:

[0110] The third calculation unit 4024 is used to count the number of adjacent nodes of each node to obtain the degree of each node; the fourth calculation unit 4025 is used to calculate the sum of the transaction amounts corresponding to each directed edge between the two nodes in each node pair as the total transaction amount of each node pair; the fifth calculation unit 4026 is used to calculate the quotient of the number of directed edges between the two nodes in each node pair and the preset time period as the transaction frequency of each node pair.

[0111] On the basis of the above embodiments, further, the corporate customer mining model is an enterprise influence model, and the enterprise influence model includes:

[0112]

[0113]

[0114] Among them, Freq(i,j) represents the transaction frequency of node pair (i,j), Amt(i,j) represents the total transaction amount of node pair (i,j), weight(i,j) represents the weight of node pair (i,j), Freq max Indicates the maximum transaction frequency of each node pair, Freq min Indicates the minimum transaction frequency of each node pair, Amt max Indicates the maximum transaction amount of each node pair, Amt min represents the minimum value of the total transaction amount of each node pair, m represents the number of adjacent nodes of the i-th node, degree(j) represents the degree of the j-th node adjacent to the i-th node, PR(i) represents the influence value of the i-th node, PR(j) represents the influence value of the j-th node adjacent to the i-th node, d is a constant, d is greater than 0 and less than 1, N represents the total number of nodes, i is a positive integer and i is less than or equal to N, j is a positive integer and j is less than or equal to m.

[0115] Figure 7 : is a schematic diagram of the structure of the excavation device for corporate customers provided by the seventh embodiment of the present invention, such as Figure 7 As shown, based on the above embodiments, the mining device for corporate customers provided in the embodiment of the present invention further includes a screening module 405, an acquisition module 406, a fourth acquisition module 407 and a recommendation module 408, wherein:

[0116] The screening module 405 is used to screen out external corporate customers and internal corporate customers from the corporate customers corresponding to each node; the acquisition module 406 is used to obtain the holding status of local financial products of each internal corporate customer; the fourth acquisition module 407 is used to obtain the interest level value of the external corporate customer in the local financial product based on the holding status of the local financial products of the internal corporate customers corresponding to each node adjacent to the node corresponding to the external corporate customer and the weight of the node pair corresponding to the node corresponding to the external corporate customer; the recommendation module 408 is used to recommend the local financial product to the external corporate customer after determining that the interest level value of the external corporate customer in local finance is greater than the recommendation threshold.

[0117] On the basis of the above embodiments, further, the fourth obtaining module 407 is specifically used for:

[0118] According to the formula Calculate and obtain the interest level value p(k,u) of external corporate customer k in local financial product u, where weight(k,j) represents the weight of the node pair (k,j), d(j,u) represents the holding status of the internal corporate customers corresponding to the j-th adjacent node of external corporate customer k in local financial product u, and m represents the number of internal corporate customers corresponding to the adjacent nodes of external corporate customer k.

[0119] The embodiments of the apparatus provided in the embodiments of the present invention can be specifically used to execute the processing flow of the above-mentioned method embodiments, and its functions are not described in detail here, and reference can be made to the detailed description of the above-mentioned method embodiments.

[0120] Figure 8 is a schematic diagram of the physical structure of an electronic device provided by an embodiment of the present invention, such as Figure 8 As shown, the electronic device may include: a processor 801, a communication interface 802, a memory 803 and a communication bus 804, wherein the processor 801, the communication interface 802 and the memory 803 complete mutual communication through the communication bus 804. The processor 801 may call the logic instructions in the memory 803 to execute the following method: based on the transaction flow of each corporate customer in a preset time period, a capital flow diagram of corporate customers is established; wherein the capital flow diagram includes each node and a directed edge connecting two nodes, wherein the node corresponds to a corporate customer one by one, and each transaction flow corresponds to a directed edge; based on each node and the edge between each node, mining parameters of corporate customers are obtained; according to the mining parameters of corporate customers and the corresponding corporate customer mining model, a mining reference value of each node is obtained; wherein the corporate customer mining model is preset; according to the mining reference value of each node and the corresponding mining threshold, a target corporate customer is obtained.

[0121] In addition, the logic instructions in the above-mentioned memory 803 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0122] The present embodiment discloses a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the methods provided by the above-mentioned method embodiments, for example, including: establishing a capital flow diagram for corporate customers based on the transaction flows of each corporate customer in a preset time period; wherein the capital flow diagram includes each node and a directed edge connecting two nodes, the nodes correspond to corporate customers one by one, and each transaction flow corresponds to a directed edge; based on each node and the edges between each node, obtaining mining parameters for corporate customers; according to the mining parameters for corporate customers and the corresponding corporate customer mining model, obtaining a mining reference value for each node; wherein the corporate customer mining model is preset; according to the mining reference value of each node and the corresponding mining threshold, obtaining the target corporate customer.

[0123] The present embodiment provides a computer-readable storage medium, which stores a computer program, and the computer program enables the computer to execute the methods provided by the above-mentioned method embodiments, for example, including: establishing a capital flow diagram for corporate customers based on the transaction flows of each corporate customer in a preset time period; wherein the capital flow diagram includes each node and a directed edge connecting two nodes, and the node corresponds to a corporate customer one by one, and each transaction flow corresponds to a directed edge; based on each node and the edge between each node, obtaining mining parameters for corporate customers; according to the mining parameters for corporate customers and the corresponding corporate customer mining model, obtaining a mining reference value for each node; wherein the corporate customer mining model is preset; according to the mining reference value of each node and the corresponding mining threshold, obtaining the target corporate customer.

[0124] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0125] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0126] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0127] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0128] In the description of this specification, the description with reference to the terms "one embodiment", "a specific embodiment", "some embodiments", "for example", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0129] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for mining corporate customers, characterized in that: include: Based on the transaction flow of each corporate customer in a preset time period, a capital flow diagram of corporate customers is established; wherein the capital flow diagram includes each node and a directed edge connecting two nodes, the nodes correspond to corporate customers one by one, and each transaction flow corresponds to a directed edge; Based on each node and the edges between nodes, obtain the mining parameters for corporate customers; According to the mining parameters of corporate customers and the corresponding corporate customer mining model, a mining reference value of each node is obtained; wherein the corporate customer mining model is preset, including a core enterprise model or an enterprise influence model; Obtain target corporate customers based on the mining reference value of each node and the corresponding mining threshold; The mining parameters of corporate customers include the out-degree and in-degree of each node, the frequency of capital inflow and outflow of each node, the total amount of capital outflow and the total amount of capital inflow of each node; accordingly, the mining parameters of corporate customers obtained based on the edges between each node include: The number of different nodes connected to each node through the capital outflow edge is counted as the out-degree of each node, and the number of different nodes connected to each node through the capital inflow edge is counted as the in-degree of each node; Calculate the quotient of the number of capital inflow edges of each node and the preset time period as the capital inflow frequency of each node, and calculate the quotient of the number of capital outflow edges of each node and the preset time period as the capital outflow frequency of each node; Calculate the sum of the transaction amounts corresponding to each capital inflow edge of each node as the total capital inflow of each node, and calculate the sum of the transaction amounts corresponding to each capital outflow edge of each node as the total capital outflow of each node; Wherein, when the corporate customer mining model is a core enterprise model, the core enterprise model includes: I(i)=In_I(i)×Out_I(i) Where I(i) represents the mining reference value, In_Degree(i) represents the in-degree of the i-th node, In_Freq(i) represents the frequency of capital inflow of the i-th node, In_Amt(i) represents the total amount of capital inflow of the i-th node, Out_Degree(i) represents the out-degree of the i-th node, Out_Freq(i) represents the frequency of capital outflow of the i-th node, Out_Amt(i) represents the total amount of capital outflow of the i-th node, In_Freq(i) represents the frequency of capital outflow of the i-th node, Out_Amt(i) represents the total amount of capital outflow of the i-th node, max Indicates the maximum value of the node's capital inflow frequency, In_Freq min Indicates the minimum frequency of capital inflow into the node, In_Amt max Indicates the maximum amount of capital inflow into the node, In_Amt min Indicates the minimum value of the total amount of funds flowing into the node, Out_Freq max Indicates the maximum value of the node's fund outflow frequency, Out_Freq min Indicates the minimum value of the node's fund outflow frequency, Out_Amt max Indicates the maximum amount of funds outflow of the node, Out_Amt min Represents the minimum value of the total amount of capital outflow of the node, N represents the total number of nodes, i is a positive integer and i is less than or equal to N.

2. The method according to claim 1, characterized in that: The mining parameters for corporate customers also include the degree of each node, the total transaction amount corresponding to each node, and the transaction frequency corresponding to each node; wherein the node pair refers to two adjacent nodes; accordingly, the mining parameters for corporate customers obtained based on the edges between each node include: Count the number of adjacent nodes of each node and obtain the degree of each node; Calculate the sum of the transaction amounts corresponding to each directed edge between the two nodes in each node pair as the total transaction amount of each node pair; The quotient of the number of directed edges between two nodes in each node pair and the preset time period is calculated as the transaction frequency of each node pair.

3. The method according to claim 2, characterized in that When the corporate customer mining model is an enterprise influence model, the enterprise influence model includes: Among them, Freq(i,j) represents the transaction frequency of node pair (i,j), Amt(i,j) represents the total transaction amount of node pair (i,j), weight(i,j) represents the weight of node pair (i,j), Freq max Indicates the maximum transaction frequency of each node pair, Freq min Indicates the minimum transaction frequency of each node pair, Amt max Indicates the maximum transaction amount of each node pair, Amt min represents the minimum value of the total transaction amount of each node pair, m represents the number of adjacent nodes of the i-th node, degree(j) represents the degree of the j-th node adjacent to the i-th node, PR(i) represents the influence value of the i-th node, PR(j) represents the influence value of the j-th node adjacent to the i-th node, d is a constant, d is greater than 0 and less than 1, N represents the total number of nodes, i is a positive integer and i is less than or equal to N, j is a positive integer and j is less than or equal to m.

4. The method according to claim 3, characterized in that Also includes: Filter out external corporate customers and internal corporate customers from the corporate customers corresponding to each node; Get the holding status of each internal corporate customer's local financial products; Obtaining the interest level value of the external corporate customer in the local financial product according to the holding status of the local financial product of the internal corporate customer corresponding to each node adjacent to the node corresponding to the external corporate customer and the weight of the node pair corresponding to the node corresponding to the external corporate customer; If it is determined that the interest level of the external corporate customer in local finance is greater than a recommendation threshold, recommending the local financial product to the external corporate customer; Wherein, obtaining the interest level value of the external corporate customer in the local financial product according to the holding status of the local financial product of the corporate customer corresponding to each node adjacent to the node corresponding to the external corporate customer and the weight of the node pair corresponding to the node corresponding to the external corporate customer includes: According to the formula Calculate and obtain the interest level value p(k,u) of external corporate customer k in local financial product u, where weight(k,j) represents the weight of the node pair (k,j), d(j,u) represents the holding status of the internal corporate customers corresponding to the j-th adjacent node of external corporate customer k in local financial product u, and m represents the number of internal corporate customers corresponding to the adjacent nodes of external corporate customer k.

5. A mining device for corporate customers, characterized in that: include: Establishing a module for establishing a capital flow diagram for corporate customers based on the transaction flow of each corporate customer in a preset time period; wherein the capital flow diagram includes each node and a directed edge connecting two nodes, the nodes correspond to corporate customers one by one, and each transaction flow corresponds to a directed edge; A first acquisition module is used to obtain mining parameters of corporate customers based on each node and the edges between each node; A second acquisition module is used to obtain a mining reference value of each node according to mining parameters of corporate customers and a corresponding corporate customer mining model; wherein the corporate customer mining model is preset, including a core enterprise model or an enterprise influence model; The third acquisition module is used to obtain the target corporate customers according to the mining reference value of each node and the corresponding mining threshold; The mining parameters for corporate customers include the out-degree and in-degree of each node, the frequency of capital inflow and the frequency of capital outflow of each node, and the total amount of capital outflow and the total amount of capital inflow of each node; accordingly, the first acquisition module includes a statistical unit, a first calculation unit and a second calculation unit, wherein: The statistical unit is used to count the number of different nodes connected to each node through the capital outflow edge as the out-degree of each node, and count the number of different nodes connected to each node through the capital inflow edge as the in-degree of each node; The first calculation unit is used to calculate the quotient of the number of capital inflow edges of each node and the preset time period as the capital inflow frequency of each node, and calculate the quotient of the number of capital outflow edges of each node and the preset time period as the capital outflow frequency of each node; The second calculation unit is used to calculate the sum of the transaction amounts corresponding to each capital inflow edge of each node as the total capital inflow amount of each node, and calculate the sum of the transaction amounts corresponding to each capital outflow edge of each node as the total capital outflow amount of each node; Wherein, when the corporate customer mining model is a core enterprise model, the core enterprise model includes: I(i)=In_I(i)×Out_I(i) Where I(i) represents the mining reference value, In_Degree(i) represents the in-degree of the i-th node, In_Freq(i) represents the frequency of capital inflow of the i-th node, In_Amt(i) represents the total amount of capital inflow of the i-th node, Out_Degree(i) represents the out-degree of the i-th node, Out_Freq(i) represents the frequency of capital outflow of the i-th node, Out_Amt(i) represents the total amount of capital outflow of the i-th node, In_Freq(i) represents the frequency of capital outflow of the i-th node, Out_Amt(i) represents the total amount of capital outflow of the i-th node, max Indicates the maximum value of the node's capital inflow frequency, In_Freq min Indicates the minimum frequency of capital inflow into the node, In_Amt max Indicates the maximum amount of capital inflow into the node, In_Amt min Indicates the minimum value of the total amount of funds flowing into the node, Out_Freq max Indicates the maximum value of the node's fund outflow frequency, Out_Freq min Indicates the minimum value of the node's fund outflow frequency, Out_Amt max Indicates the maximum amount of funds outflow of the node, Out_Amt min Represents the minimum value of the total amount of capital outflow of the node, N represents the total number of nodes, i is a positive integer and i is less than or equal to N.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 4 are implemented.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.

8. A computer program product, comprising a computer program stored on a non-transitory computer-readable storage medium, wherein the computer program comprises program instructions, and when the program instructions are executed by a computer, the computer executes the method according to any one of claims 1 to 4.

Images