Method, apparatus, electronic device, and storage medium for identifying customer transaction background

Abstract

The present application provides a method, device, electronic device and storage medium for identifying the customer transaction background. After determining the start node and end node of the customer to be identified in the target transaction relationship graph according to the transaction information of the customer to be identified, based on the start node and the end node, the weight value of each path that meets the first condition is determined in the target transaction relationship graph; the first condition is that the start and end points of the path are the start node and the end node respectively, and the number of nodes passed by the path is less than or equal to the preset node upper limit; finally, according to the weight values of each path that meets the first condition, the total probability of the funds reflux of the customer to be identified is calculated. That is, this case can obtain the potential and deep-level transaction links of customers based on the transaction relationship graph, replacing the single artificial subjective judgment management method, and by advancing the risk management, strengthening the authenticity review of the customer transaction background before lending, reducing the risk of funds being misappropriated and difficult to recover from the source.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a method, device, electronic device and storage medium for identifying the customer transaction background. Background Art

[0002] The overall credit management process of banks is generally divided into three links: pre-loan investigation, in-loan approval, and post-loan monitoring. In the pre-loan link, the customer manager is responsible for conducting due diligence on the customer's situation and forming application materials. In the in-loan link, the approval personnel will approve the business based on the materials provided by the customer manager.

[0003] Due to the influence of subjective factors of the investigators, it is easy to "beautify" the customer transaction background, making it difficult for the approval personnel to accurately grasp the actual needs of the customers, resulting in the misappropriation of funds after the loan is released. In the post-loan monitoring link, although there are many monitoring means, there is still a situation where the credit funds are mixed with the enterprise's own funds after being released, and the enterprise usually conceals its illegal disbursement behavior in various ways, making it difficult for the post-loan management personnel to monitor; and even if the management personnel require the enterprise to return immediately after discovering the enterprise's illegal disbursement behavior, it is difficult to recover according to the rectification requirements due to the misappropriation of the credit funds. Summary of the Invention

[0004] In view of this, the present application provides a method, device, electronic device and storage medium for identifying the customer transaction background, which iteratively calculates and simulates the potential and deep transaction links of the customer based on the transaction relationship graph, replaces the single artificial subjective judgment management method, improves the objectivity of customer access, and strengthens the authenticity review of the customer transaction background before the loan by preposing the risk management, reducing the risk of misappropriation and difficult recovery of funds from the source.

[0005] To achieve the above object, the embodiments of the present invention provide the following technical solutions:

[0006] The first aspect of the present invention discloses a method for identifying the customer transaction background, including:

[0007] Determine the start node and end node of the to-be-identified customer in the target transaction relationship graph according to the transaction information of the to-be-identified customer;

[0008] Based on the start node and the end node, determine the weight value of each path that meets the first condition in the target transaction relationship graph; the first condition is that the start and end points of the path are the start node and the end node respectively, and the number of nodes passed by the path is less than or equal to the preset node upper limit;

[0009] Calculate according to the weight value of each path that meets the first condition to obtain the total probability of the capital return of the to-be-identified customer.

[0010] Optionally, in the above method for identifying the customer transaction background, after determining the start node and the end node of the customer to be identified in the target transaction relationship graph according to the customer transaction information of the customer to be identified, it further includes:

[0011] Based on the start node and the end node, determine the weight value of each path that meets the second condition in the target transaction relationship graph; the second condition is that the start point of the path is the start node, the end point of the path is not the end node, and the number of path nodes is greater than the preset upper limit of nodes;

[0012] Analyze according to the weight value of each path that meets the second condition to obtain the transaction result of the agreed payment object of the customer to be identified.

[0013] Optionally, in the above method for identifying the customer transaction background, it further includes:

[0014] Generate corresponding prompt information according to the total probability of capital return of the customer to be identified and / or the transaction result of the agreed payment object of the customer to be identified, and push the prompt information.

[0015] Optionally, in the above method for identifying the customer transaction background, determining the start node and the end node of the customer to be identified in the target transaction relationship graph according to the transaction information of the customer to be identified includes:

[0016] Obtain the payment object enterprise and the loan application enterprise in the transaction information of the customer to be identified;

[0017] Respectively use the payment object enterprise as the start node and the loan application enterprise as the end node.

[0018] Optionally, in the above method for identifying the customer transaction background, the generation process of the target transaction relationship graph includes:

[0019] Obtain the historical transaction data of the customer to be identified;

[0020] Create an initial transaction relationship graph of the customer to be identified according to the historical transaction data;

[0021] Obtain the investment relationship data and the guarantee relationship data of the customer to be identified, and adjust the initial transaction relationship graph according to the investment relationship data and the guarantee relationship data to obtain the target transaction relationship graph.

[0022] Optionally, in the above method for identifying the customer transaction background, creating an initial transaction relationship graph of the customer to be identified according to the historical transaction data includes:

[0023] Determine the transaction object and the fund flow direction of each transaction in the historical transaction data respectively;

[0024] Using the transaction object as the customer node, connect each of the customer nodes respectively according to the fund flow direction of each transaction object, and set weights for each transaction object according to the transaction amount and transaction frequency of each transaction, and create an initial transaction relationship graph.

[0025] Optionally, in the above method for identifying the customer transaction background, adjusting the initial transaction relationship graph according to the investment relationship data and the guarantee relationship data to obtain the target transaction relationship graph includes:

[0026] Determine the adjustment order of the investment relationship data and the guarantee relationship data;

[0027] According to the adjustment order, adjust the initial transaction relationship graph in sequence to obtain the target transaction relationship graph; wherein, the adjustment order of the investment relationship data precedes the adjustment order of the guarantee relationship data.

[0028] A second aspect of the present invention discloses a device for identifying the customer transaction background, including:

[0029] A first determination unit, configured to determine a start node and an end node of the to-be-identified customer in the target transaction relationship graph according to the transaction information of the to-be-identified customer;

[0030] A second determination unit, configured to determine the weight value of each path that meets the first condition in the target transaction relationship graph based on the start node and the end node; the first condition is that the start and end points of the path are the start node and the end node respectively, and the number of nodes passed by the path is less than or equal to a preset node upper limit;

[0031] A first calculation unit, configured to calculate according to the weight value of each path that meets the first condition to obtain the total probability of the fund return of the to-be-identified customer.

[0032] A third aspect of the present invention discloses an electronic device, including: a processor and a memory, wherein:

[0033] The memory is used to store computer instructions;

[0034] The processor is configured to execute the computer instructions stored in the memory, and specifically execute the method for identifying the customer transaction background as described in any one of the first aspect.

[0035] A fourth aspect of the present invention discloses a storage medium, used to store a program, and when the program is executed, it is used to execute the method for identifying the customer transaction background as described in any one of the first aspect.

[0036] The customer transaction background identification method provided by the present invention first determines the starting node and the ending node of the customer to be identified in the target transaction relationship graph according to the transaction information of the customer to be identified; then, based on the starting node and the ending node, determines the weight value of each path that meets the first condition in the target transaction relationship graph; the first condition is that the starting and ending points of the path are the starting node and the ending node respectively, and the number of nodes passed by the path is less than or equal to the preset upper limit of nodes; finally, calculates based on the weight value of each path that meets the first condition to obtain the total probability of the capital return of the customer to be identified. That is to say, this solution can iteratively calculate and simulate the potential and deep transaction links of the customer based on the transaction relationship graph, replacing the single artificial subjective judgment management method, improving the objectivity of customer access, and strengthening the authenticity review of the customer transaction background before the loan by advancing the risk management, reducing the risk hidden dangers of misappropriation and difficult recovery of funds from the source. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0038] Figure 1 It is a flowchart of a customer transaction background identification method provided by an embodiment of the present application;

[0039] Figure 2 It is a flowchart of the generation of a target transaction relationship graph provided by an embodiment of the present application;

[0040] Figure 3 It is a flowchart of the generation of an initial transaction relationship graph provided by an embodiment of the present application;

[0041] Figure 4 It is a specific flowchart of the generation of a target transaction relationship graph provided by an embodiment of the present application;

[0042] Figure 5 It is a flowchart of the determination of the starting node and the ending node provided by an embodiment of the present application;

[0043] Figure 6 It is a flowchart of the determination of the path that meets the conditions and its weight provided by an embodiment of the present application;

[0044] Figure 7 It is a flowchart of another customer transaction background identification method provided by an embodiment of the present application;

[0045] Figure 8 It is a flowchart of another method for identifying the customer transaction background provided by an embodiment of the present application;

[0046] Figure 9 It is a schematic structural diagram of a device for identifying the customer transaction background provided by an embodiment of the present application. Detailed implementation manners

[0047] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0048] First of all, it should be noted that through the research of the inventor, it is found that at present, the pre-loan customer transaction background investigation can only be carried out by manual investigation, with slow investigation speed and high labor cost.

[0049] In response to this, an embodiment of the present application provides a method for identifying the customer transaction background, which can iteratively calculate and simulate the potential and deep transaction links of the customer based on the transaction relationship graph, replacing the single manual subjective judgment management method, improving the objectivity of customer access, and strengthening the authenticity review of the pre-loan customer transaction background by advancing risk management, thereby reducing the risk hidden dangers of misappropriation and difficult recovery of funds from the source.

[0050] Please refer to Figure 1 , the method for identifying the customer transaction background mainly includes the following steps:

[0051] S100. Determine the start node and end node of the customer to be identified in the target transaction relationship graph according to the transaction information of the customer to be identified.

[0052] Among them, the customer to be identified is a customer who needs to perform customer transaction background identification but has not been identified yet. Specifically, the customer to be identified is generally the same as the loan application enterprise in the transaction information.

[0053] It should be noted that, please refer to Figure 2 , the generation process of the target transaction relationship graph can be as follows:

[0054] S300. Obtain the historical transaction data of the customer to be identified.

[0055] In practical applications, the historical transaction data of the customer to be identified can be obtained by accessing the cloud database to obtain the historical transaction data stored in the accounting system.

[0056] Of course, the method for obtaining the historical transaction data of the customer to be identified is not limited to this. The historical transaction data of the customer to be identified can also be obtained through other existing methods. The present application does not limit the specific method of obtaining, and all are within the protection scope of the present application.

[0057] S302. Create an initial transaction relationship graph of the customer to be identified according to the historical transaction data.

[0058] In practical applications, combined with Figure 4 , when performing step S302, the specific process of creating an initial transaction relationship graph of the customer to be identified according to the historical transaction data can be as Figure 3 shown and may include the following steps:

[0059] S400. Respectively determine the transaction object and the fund flow direction of each transaction in the historical transaction data.

[0060] Among them, through data processing, the transaction object and the fund flow direction of each transaction in the historical transaction data can be summarized, so as to determine the transaction object and the fund flow direction of each transaction in the historical transaction data.

[0061] It should be noted that in order to improve the data processing speed, the historical transaction data can be cleaned before determining the transaction object and the fund flow direction of each transaction in the historical transaction data, so as to eliminate invalid data such as loan disbursement and repayment.

[0062] S402. Use the transaction object as the customer node, connect each customer node according to the fund flow direction of each transaction object respectively, and set weights for each transaction object according to the transaction amount and transaction frequency of each transaction, so as to create an initial transaction relationship graph.

[0063] Among them, using the transaction object as the customer node, the direction pointed by the arrow is the fund transfer direction. First, connect each transaction object according to the fund flow direction, and then for each transaction object, set weights for each transaction object according to the transaction amount and transaction frequency of each transaction of each transaction object, so as to obtain an initial transaction relationship graph created with key information such as transaction objects, fund flow directions and transaction amounts in the historical transaction data.

[0064] S304. Obtain the investment relationship data and guarantee relationship data of the customer to be identified, and adjust the initial transaction relationship graph according to the investment relationship data and guarantee relationship data to obtain the target transaction relationship graph.

[0065] In practical applications, the adjustment order of the investment relationship data and the guarantee relationship data for the initial transaction relationship graph can be determined first, and then the initial transaction relationship graph can be adjusted in turn according to the adjustment order to obtain the target transaction relationship graph. Among them, the adjustment order of the investment relationship data can be prior to that of the guarantee relationship data; of course, this is not limited thereto, and it can also be determined according to the specific application environment and the user. Regardless of the adjustment order, it is within the protection scope of this application.

[0066] Combined with Figure 4 , taking the adjustment order of the investment relationship data being prior to that of the guarantee relationship data as an example, the specific process of adjusting the initial transaction relationship graph according to the investment relationship data and the guarantee relationship data to obtain the target transaction relationship graph can be as follows:

[0067] In the process of adjusting the initial transaction relationship graph according to the investment relationship data, the equity relationship of each trading object can be penetrated according to the investment relationship data first to obtain the shareholding ratio of each shareholder of each trading object. Then, the connection relationship of each customer node in the initial transaction relationship graph can be supplemented according to the shareholding ratio of each shareholder of each trading object; among them, the investment relationship can be approximately regarded as a two-way transaction relationship. Finally, according to the shareholding ratio and historical transaction amount of each trading node, the weight value corresponding to each edge starting from the corresponding customer node is adjusted.

[0068] In practical applications, the weight value of the corresponding edge can be amplified by combining the investment amount and shareholding ratio of each customer node, and the remaining weights are distributed to other edges according to the original ratio. Among them, the adjustment rule for re-adjusting the weight value according to the shareholding ratio can be:

[0069] If the holding ratio is 100%, the adjusted weight value is 2. Among them, if the original weight is between 0.5 and 0.8, the adjusted weight value is 0.8; if the original weight is greater than 0.8, the adjusted weight value is 1.

[0070] If the holding ratio is 80% - 100%, the adjusted weight value is 1.5. Among them, if the original weight is between 0.67 and 0.8, the adjusted weight value is 0.8; if the original weight is greater than 0.8, the adjusted weight value is 1.

[0071] If the holding ratio is 51% - 79%, the adjusted weight value is 1.34. Among them, if the original weight is greater than 0.75, the adjusted weight value is 0.8; if the original weight is greater than 0.8, the adjusted weight value is 1.

[0072] The weight value of the remaining customer nodes is adjusted to 1 - the adjusted weight value of this edge * the original weight value i / ∑(the original weight value i).

[0073] After adjusting the initial transaction relationship graph according to the investment relationship data, the guarantee relationship data can be processed first. The guarantee amount of each customer node is counted and it is judged whether there is a guarantee circle to obtain a judgment result. According to the counted guarantee amount of each customer node and the judgment result, the connection relationship between customer nodes is supplemented. Among them, the guarantee relationship can be approximately regarded as a two-way transaction relationship. Finally, according to the guarantee amount of each customer node, the historical transaction amount, and the judgment result of whether there is a guarantee circle, the weight value corresponding to each edge starting from the corresponding customer node is adjusted, and the initial transaction relationship graph adjusted according to the investment relationship data and the guarantee relationship data is used as the target transaction relationship graph.

[0074] In practical applications, it can be judged whether there is a guarantee circle according to the starting node and the ending node. Among them, the upper limit of the number of nodes in the circle can be set according to the requirements of each financial institution, and the circular chain with mutual guarantee relationships composed of no more than the set number of nodes in the circle is the guarantee circle.

[0075] Combined with the above, it can be understood that for investment relationships, the penetration principle can be adopted to restore the complete investment and financing relationships to the greatest extent and achieve the first correction of the initial transaction relationship graph. Similarly, for guarantee relationships, it is necessary to first analyze whether there is a guarantee circle and use the judgment result as an influencing factor for weight value setting to achieve the second correction of the initial transaction relationship graph.

[0076] It should be noted that during the process of adjusting the initial transaction relationship graph by introducing investment relationship data and guarantee relationship data, no new customer nodes are added additionally. Only the transaction relationships and weight settings are supplemented and adjusted based on the existing customer nodes.

[0077] It also needs to be noted that the investment relationship data is external industrial and commercial data and can be obtained through the corresponding industrial and commercial system; the guarantee relationship data is credit information data and can be obtained through the corresponding credit information system. Of course, the specific methods for obtaining investment relationship data and guarantee relationship data are not limited to the above, and can also be obtained through other existing methods. This application does not make specific limitations on them, and they are all within the protection scope of this application.

[0078] In practical applications, the specific implementation process of step S100, determining the starting node and the ending node of the to-be-identified customer in the target transaction relationship graph according to the transaction information of the to-be-identified customer, can be as Figure 5 shown, and mainly includes the following steps:

[0079] S200. Obtain the payment object enterprise and the loan application enterprise in the transaction information of the to-be-identified customer.

[0080] In practical applications, the transaction information of the customer to be identified is usually uploaded by the customer manager. The transaction information generally includes: the enterprise applying for a loan, the payment date, the payment amount, and the enterprise receiving the payment. Among them, the enterprise applying for a loan is generally composed of a customer number and an enterprise name.

[0081] It should be noted that the enterprise receiving the payment and the enterprise applying for a loan in the transaction information can be obtained through the transaction information uploaded by the customer manager. Of course, this is not the only way. The specific methods for obtaining the enterprise receiving the payment and the enterprise applying for a loan in the transaction information of the customer to be identified can also refer to other existing technologies, which will not be elaborated in this application and are all within the scope of protection of this application.

[0082] S202: Respectively take the enterprise receiving the payment as the starting node and the enterprise applying for a loan as the ending node.

[0083] In practical applications, after obtaining the enterprise receiving the payment and the enterprise applying for a loan, the enterprise receiving the payment can be respectively taken as the starting node and the enterprise applying for a loan as the ending node.

[0084] S102: Based on the starting node and the ending node, determine the weight value of each path that meets the first condition in the target transaction relationship graph.

[0085] Among them, the first condition is that the start and end points of the path are the starting node and the ending node respectively, and the number of nodes passed by the path is less than or equal to the preset node upper limit.

[0086] In practical applications, all feasible paths between the starting node and the ending node can be traversed in the target transaction relationship graph, and the possibility of each path occurring can be calculated to obtain the weight value of each path whose number of nodes passed by is less than or equal to the preset node upper limit.

[0087] Since each node in the target transaction relationship graph is based on an enterprise (transaction object), the direction of the arrow points to the direction of capital flow, and the weight value is based on a single node, the weight value of each path can be obtained by dividing the sum of the total amount flowing from this node to another node multiplied by the first influence factor plus the total number of transactions multiplied by the second influence factor by the sum of the total amount flowing out from this node to all nodes multiplied by the first influence factor plus the total number of transactions multiplied by the second influence factor. Among them, the sum of the first influence factor and the second influence factor is 1.

[0088] Suppose the total debit settlement volume of enterprise A is X, the number of settlement transactions is Y, the total amount flowing from enterprise A to enterprise B is X1, the total number of transactions is Y1, the first influence factor is U1, and the second influence factor is U2. Then the weight from enterprise A to enterprise B is: X1 / (X*U1)+Y1 / (Y*U2).

[0089] In practical applications, the first influence factor and the second influence factor are generally determined based on the analysis of historical related transactions. Generally, both the first influence factor and the second influence factor are approximately 0.5.

[0090] It should be noted that in combination with Figure 6 , the specific process of determining the weight value of each path that meets the first condition in the target transaction relationship graph can be as follows:

[0091] First, input the transaction information of the customer to be identified. Then, set the upper limit of the number of nodes passed by the path. Next, locate the starting node. Then, calculate the next flow direction of the funds based on the starting node and determine whether there is a next connectable node. If it is determined that there is a next connectable node, the number of nodes passed by the path is incremented by 1 and it is determined whether the number of nodes passed by the path reaches the upper limit of the number of nodes passed by the path. If it is determined that the number of nodes passed by the path does not reach the upper limit of the number of nodes passed by the path, it is determined whether a loop is formed with the termination node. If it is determined that a loop is formed with the termination node, output the fund transfer path existing between the starting node and the termination node and the corresponding weight value.

[0092] S104. Calculate based on the weight value of each path that meets the first condition to obtain the total probability of fund reflux of the customer to be identified.

[0093] In practical applications, the weight values of each path that meets the first condition can be directly summed to obtain the total probability of fund reflux of the customer to be identified. Among them, the weight value of each path can be obtained by multiplying the weight values of each segment of the path that makes up the path.

[0094] It can be understood that the total probability of fund reflux of the customer to be identified is the total probability that the funds of the payment object agreed by the customer to be identified flow back to itself.

[0095] It should be noted that in practical applications, sensitive enterprises can also be marked according to the target transaction relationship graph, that is Figure 4 as shown, the real estate enterprises are marked.

[0096] It also should be noted that the target transaction relationship graph can be a weighted directed graph, which stores data in the form of a graph structure. Multiple customer nodes are connected by directed edges, and each edge has a corresponding weight value. The weight is determined by historical trading volumes and other related relationships. The greater the weight, the higher the probability of the transaction occurring.

[0097] Based on the above principle, for the method for identifying the customer transaction background provided in this embodiment, first, according to the transaction information of the customer to be identified, the starting node and the ending node of the customer to be identified in the target transaction relationship graph are determined; then, based on the starting node and the ending node, the weight value of each path that meets the first condition is determined in the target transaction relationship graph; the first condition is that the start and end points of the path are the starting node and the ending node respectively, and the number of nodes passed by the path is less than or equal to the preset node upper limit; finally, according to the weight values of each path that meets the first condition, the total probability of the customer's capital reflux is calculated, that is, this solution can iteratively calculate and simulate the potential and deep transaction links of the customer based on the transaction relationship graph, replacing the single artificial subjective judgment management method, improving the objectivity of customer access, and by preposing risk management, strengthening the authenticity review of the customer transaction background before loan, reducing the risk of fund misappropriation and difficult recovery from the source.

[0098] In addition, this application can also accurately and quickly sort out and summarize the associated relationship information of customers, reduce the links of manual information collection and individual access, and effectively improve the approval efficiency.

[0099] Optionally, in another embodiment provided by this application, after performing step S100, according to the transaction information of the customer to be identified, determining the starting node and the ending node of the customer to be identified in the target transaction relationship graph, please refer to Figure 7 , the method for identifying the customer transaction background further includes:

[0100] S500. Based on the starting node and the ending node, determine the weight value of each path that meets the second condition in the target transaction relationship graph.

[0101] Among them, the second condition is that the starting point of the path is the starting node, the ending point of the path is not the ending node, and the number of path nodes is greater than the preset node upper limit.

[0102] In practical applications, all feasible paths of the starting node can be traversed in the target transaction relationship graph, and the possibility of each path occurring can be calculated to obtain the weight value of each path that meets the starting node, does not form a loop with the ending node, and the number of path nodes is greater than the preset node upper limit.

[0103] It should be noted that, also in combination with Figure 6 , the specific process of determining the weight value of each path that meets the second condition in the target transaction relationship graph can be as follows:

[0104] First, input the transaction information of the customer to be identified. Then, set the upper limit of the number of nodes passed through in the path. Next, locate the starting node. Then, calculate the next flow direction of the funds based on the starting node and determine whether there is a next connectable node. If it is determined that there is a next connectable node, the number of nodes passed through in the path is incremented by 1 and it is determined whether the number of nodes passed through in the path reaches the upper limit of the number of nodes passed through in the path. If it is determined that the number of nodes passed through in the path does not reach the upper limit of the number of nodes passed through in the path, it is determined whether a loop is formed with the termination node. If it is determined that no loop is formed with the termination node, the next flow direction of the funds is calculated and it is determined whether there is a next connectable node. If it is determined that there is no next connectable node, the fund transfer path of the starting node and the corresponding weight value are directly output. If it is determined that the number of nodes passed through in the path reaches the upper limit of the number of nodes passed through in the path, the fund transfer path of the starting node and the corresponding weight value are also directly output.

[0105] S502. Analyze according to the weight value of each path that meets the second condition to obtain the transaction result of the agreed payment object of the customer to be identified.

[0106] In practical applications, each path that meets the second condition represents a path that does not connect the starting node and the termination node. It is also necessary to check whether the funds flow to prohibited areas. Therefore, it is necessary to sort out the main flow directions and analyze the actual production and operation conditions of the agreed payment object.

[0107] It should be noted that the common transaction paths of the agreed payment object can be output, and based on the common transaction paths, it can be judged whether there is an indirect flow to prohibited areas in its general fund trend, so as to obtain the transaction result of the agreed payment object of the customer to be identified.

[0108] Optionally, in another embodiment provided by the present application, please refer to Figure 8 , the method for identifying the customer transaction background further includes:

[0109] S600. Generate corresponding prompt information according to the total probability of fund return of the customer to be identified and / or the transaction result of the agreed payment object of the customer to be identified, and push the prompt information.

[0110] In practical applications, the probability that the funds of the agreed payment object of the customer to be identified return to itself and the corresponding path can be obtained from the total probability of fund return of the customer to be identified. The historical fund flow of the agreed payment object of the customer to be identified and its associated enterprise relationship can be obtained from the agreed payment transaction result of the customer to be identified. Then, the two are combined and analyzed to generate corresponding prompt information.

[0111] Among them, the prompt information generated according to the total probability of the capital return of the customer to be identified may include: all the capital flow path diagrams between the customer to be identified and the agreed payment object, the weight value corresponding to each path, and the total probability of the capital return of the customer to be identified; while the prompt information generated according to the transaction result of the agreed payment object of the customer to be identified may include: the path and probability value that flow into the prohibited area, and the historical transaction customers of the agreed payment object and the existing association relationship. Among them, the display quantity of the historical transaction customers of the agreed payment object can be determined according to the specific application environment and user requirements. For example, only the top 5 customers with the highest transaction frequency are displayed. Of course, it is not limited to this. No matter how it is displayed, it is within the protection scope of this application.

[0112] It should be noted that after generating the corresponding prompt information, the generated prompt information can be pushed to the corresponding credit approval personnel for reference by the credit approval personnel.

[0113] Based on the customer transaction background identification method provided in the above embodiment, another embodiment of this application also provides a customer transaction background identification device. Please refer to Figure 9 , and this device mainly includes:

[0114] The first determination unit 100 is used to determine the starting node and the ending node of the customer to be identified in the target transaction relationship graph according to the transaction information of the customer to be identified.

[0115] The second determination unit 102 is used to determine the weight value of each path that meets the first condition in the target transaction relationship graph based on the starting node and the ending node; the first condition is that the start and end points of the path are the starting node and the ending node respectively, and the number of nodes passed by the path is less than or equal to the preset node upper limit.

[0116] The first calculation unit 104 is used to calculate according to the weight value of each path that meets the first condition to obtain the total probability of the capital return of the customer to be identified.

[0117] Optionally, in the above customer transaction background identification device, it further includes:

[0118] The third determination unit is used to determine the weight value of each path that meets the second condition in the target transaction relationship graph based on the starting node and the ending node; the second condition is that the starting point of the path is the starting node, the ending point of the path is not the ending node, and the number of path nodes is greater than the preset node upper limit.

[0119] The first analysis unit is used to analyze according to the weight value of each path that meets the second condition to obtain the transaction result of the agreed payment object of the customer to be identified.

[0120] Optionally, in the above customer transaction background identification device, it further includes:

[0121] A generating and pushing unit, configured to generate a corresponding prompt message according to the total probability of capital return of the customer to be identified and / or the transaction result of the agreed payment object of the customer to be identified, and push the prompt message.

[0122] Optionally, in the above customer transaction background identification device, the first determining unit 100 is specifically configured to:

[0123] Obtain the payment object enterprise and the loan application enterprise in the transaction information of the customer to be identified.

[0124] Respectively use the payment object enterprise as the starting node and the loan application enterprise as the ending node.

[0125] Optionally, in the above customer transaction background identification device, the generation process of the target transaction relationship graph includes:

[0126] Obtain the historical transaction data of the customer to be identified.

[0127] Create an initial transaction relationship graph of the customer to be identified according to the historical transaction data.

[0128] Obtain the investment relationship data and the guarantee relationship data of the customer to be identified, and adjust the initial transaction relationship graph according to the investment relationship data and the guarantee relationship data to obtain the target transaction relationship graph.

[0129] Optionally, in the above customer transaction background identification device, creating an initial transaction relationship graph of the customer to be identified according to the historical transaction data includes:

[0130] Respectively determine the transaction object and the capital flow direction of each transaction in the historical transaction data.

[0131] Use the transaction object as the customer node, connect each customer node according to the capital flow direction of each transaction object respectively, and set weights for each transaction object according to the transaction amount and transaction frequency of each transaction, so as to create an initial transaction relationship graph.

[0132] Optionally, in the above customer transaction background identification device, adjusting the initial transaction relationship graph according to the investment relationship data and the guarantee relationship data to obtain the target transaction relationship graph includes:

[0133] Determine the adjustment order of the investment relationship data and the guarantee relationship data.

[0134] According to the adjustment order, adjust the initial transaction relationship graph in sequence to obtain the target transaction relationship graph; wherein, the adjustment order of the investment relationship data is prior to the adjustment order of the guarantee relationship data.

[0135] The customer transaction background recognition device provided in this embodiment includes: a first determination unit 100 configured to determine a start node and an end node of a customer to be recognized in a target transaction relationship graph according to the transaction information of the customer to be recognized; a second determination unit 102 configured to determine the weight value of each path that meets the first condition in the target transaction relationship graph based on the start node and the end node; the first condition is that the start and end points of the path are the start node and the end node respectively, and the number of nodes passed by the path is less than or equal to a preset node upper limit; a first calculation unit 104 configured to calculate based on the weight values of each path that meets the first condition to obtain the total probability of the customer's fund reflux. That is, this solution can iteratively calculate and simulate the potential and deep transaction links of the customer based on the transaction relationship graph, replacing the single artificial subjective judgment management method, improving the objectivity of customer access, and strengthening the authenticity review of the customer's transaction background before loan by advancing risk management, reducing the risk of fund misappropriation and difficult recovery from the source.

[0136] Optionally, another embodiment of the present application further provides an electronic device, including: a processor and a memory, where:

[0137] The memory is used to store computer instructions;

[0138] The processor is configured to execute the computer instructions stored in the memory, and specifically execute the customer transaction background recognition method described in any of the above embodiments.

[0139] It should be noted that the relevant descriptions of the customer transaction background recognition method can be referred to Figures 1 to 8 the corresponding embodiments, and will not be elaborated here.

[0140] Optionally, another embodiment of the present application further provides a storage medium for storing a program, and when the program is executed, it is used to implement the customer transaction background recognition method described in any of the above embodiments.

[0141] It should be noted that the relevant descriptions of the customer transaction background recognition method can be referred to Figures 1 to 8 the corresponding embodiments, and will not be elaborated here.

[0142] The features described in the various embodiments in this specification can be replaced or combined with each other. For the same or similar parts among the various embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for a system or system embodiment, since it is basically similar to the method embodiment, the description is relatively simple. For the relevant parts, reference can be made to the partial description of the method embodiment. The systems and system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without creative efforts.

[0143] Those skilled in the art can further realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0144] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the broadest scope consistent with the principles and novel features disclosed herein.

[0145] It should also be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising said element.

Claims

1. A method for identifying a customer transaction background, characterized in that, Including: Determine the start node and end node of the to-be-identified customer in the target transaction relationship graph according to the transaction information of the to-be-identified customer; Based on the start node and the end node, determine the weight value of each path that meets the first condition in the target transaction relationship graph; the first condition is that the start and end points of the path are the start node and the end node respectively, and the number of nodes passed by the path is less than or equal to the preset node upper limit; Calculate according to the weight values of each path that meets the first condition to obtain the total probability of the capital return of the to-be-identified customer; Based on the start node and the end node, determine the weight value of each path that meets the second condition in the target transaction relationship graph; the second condition is that the start point of the path is the start node, the end point of the path is not the end node, and the number of path nodes is greater than the preset node upper limit; Analyze according to the weight values of each path that meets the second condition to obtain the transaction result of the agreed payment object of the to-be-identified customer; The generation process of the target transaction relationship graph includes: Obtain the historical transaction data of the to-be-identified customer; Respectively determine the transaction object and the capital flow direction of each transaction in the historical transaction data; Use the transaction object as the customer node, connect each customer node according to the capital flow direction of each transaction object respectively, and set weights for each transaction object according to the transaction amount and transaction frequency of each transaction to create an initial transaction relationship graph; Obtain the investment relationship data and guarantee relationship data of the to-be-identified customer, and determine the adjustment order of the investment relationship data and the guarantee relationship data; According to the adjustment order, adjust the initial transaction relationship graph in turn to obtain the target transaction relationship graph; among them, the adjustment order of the investment relationship data is prior to the adjustment order of the guarantee relationship data.

2. The method for identifying a customer transaction background according to claim 1, characterized in that, Also including: Generate corresponding prompt information according to the total probability of the capital return of the to-be-identified customer and / or the transaction result of the agreed payment object of the to-be-identified customer, and push the prompt information.

3. The method for identifying a customer transaction background according to any one of claims 1-2, characterized in that, Determine the start node and end node of the to-be-identified customer in the target transaction relationship graph according to the transaction information of the to-be-identified customer, including: Obtain the payment object enterprise and the loan application enterprise in the transaction information of the to-be-identified customer; Respectively use the payment object enterprise as the start node and the loan application enterprise as the end node.

4. A device for identifying a customer transaction background, characterized in that, Including: The first determination unit is used to determine the start node and end node of the to-be-identified customer in the target transaction relationship graph according to the transaction information of the to-be-identified customer; The second determination unit is used to determine the weight value of each path that meets the first condition in the target transaction relationship graph based on the start node and the end node; the first condition is that the start and end points of the path are the start node and the end node respectively, and the number of nodes passed by the path is less than or equal to the preset node upper limit; The first calculation unit is used to calculate according to the weight values of each path that meets the first condition to obtain the total probability of the capital return of the to-be-identified customer; A third determination unit, configured to determine the weight value of each path that meets the second condition in the target transaction relationship graph based on the starting node and the ending node; the second condition is that the starting point of the path is the starting node, the ending point of the path is not the ending node, and the number of path nodes is greater than the preset upper limit of nodes; A first analysis unit, configured to analyze according to the weight value of each path that meets the second condition to obtain the transaction result of the agreed payment object of the customer to be identified; In the recognition device of the customer transaction background, the generation process of the target transaction relationship graph includes: Obtain the historical transaction data of the customer to be identified; Respectively determine the transaction object and the capital flow direction of each transaction in the historical transaction data; Using the transaction object as the customer node, respectively connect each customer node according to the capital flow direction of each transaction object, and set weights for each transaction object according to the transaction amount and transaction frequency of each transaction, and create an initial transaction relationship graph; Obtain the investment relationship data and the guarantee relationship data of the customer to be identified, and determine the adjustment order of the investment relationship data and the guarantee relationship data; According to the adjustment order, adjust the initial transaction relationship graph in sequence to obtain the target transaction relationship graph; wherein, the adjustment order of the investment relationship data is prior to the adjustment order of the guarantee relationship data.

5. An electronic device, characterized in that, Including: A processor and a memory, wherein: The memory is used to store computer instructions; The processor is configured to execute the computer instructions stored in the memory, and specifically execute the customer transaction background recognition method according to any one of claims 1-3.

6. A storage medium, characterized in that, For storing a program, when the program is executed, it is used to execute the customer transaction background recognition method according to any one of claims 1-3.

Images