Customer Transaction Risk Control Method and System for Bank Branches

Abstract

The present invention provides a method and system for controlling customer transaction risks at bank branches, which relates to the technical field of transaction data processing. The method includes: classifying transactions according to the transaction data; determining the transaction category vector corresponding to the number at the current moment; determining the transaction category vector corresponding to the number at the historical moment; determining the similar historical moment at the current moment according to the multiple transaction category vectors corresponding to the current moment and the multiple transaction category vectors corresponding to each historical moment among the multiple historical moments; determining the main transaction category of the bank branch and the risk control model corresponding to the main transaction category according to the transaction data after the similar historical moment, and sending the main transaction category and the risk control model to the bank branch; when a customer conducts a transaction at the bank branch, controlling the risk of the customer's current transaction according to the main transaction category of the bank branch and the risk control model.

Description

Technical Field

[0001] The present invention relates to the technical field of transaction data processing, and particularly to a method and system for controlling customer transaction risks at bank branches. Background Art

[0002] This section aims to provide background or context for the embodiments of the present invention described in the claims. The descriptions herein are not admitted to be prior art merely because they are included in this section.

[0003] Currently, the control of transaction risks at bank branches mainly relies on the experience accumulation of business personnel and lacks the support of actual data, resulting in relatively lagging transaction risk management and being unable to timely and accurately detect the risks in bank operations.

[0004] In summary, there is an urgent need for a technical solution that can overcome the above defects and can timely detect the transaction risks at bank branches. Summary of the Invention

[0005] To solve the problems existing in the prior art, the present invention proposes a method and system for controlling customer transaction risks at bank branches.

[0006] In the first aspect of the embodiments of the present invention, a method for controlling customer transaction risks at bank branches is proposed, including:

[0007] Obtain the transaction data of the bank branch, classify the transactions according to the transaction data, and obtain multiple transaction categories;

[0008] Number the time within a set period at the current moment. For each number, determine the transaction category vector corresponding to the current moment according to the transaction data of the bank branch at the time corresponding to the number;

[0009] For multiple historical moments, number the time within a set period at the historical moment. For each number, determine the transaction category vector corresponding to the historical moment according to the transaction data of the bank branch at the time corresponding to the number;

[0010] Determine the similar historical moment of the current moment according to the multiple transaction category vectors corresponding to the current moment and the multiple transaction category vectors corresponding to each historical moment among the multiple historical moments;

[0011] Determine the main transaction category of the bank branch and the risk control model corresponding to the main transaction category according to the transaction data after the similar historical moment, and send the main transaction category and the risk control model to the bank branch;

[0012] When a customer conducts a transaction at the bank branch, control the risk of the customer's current transaction according to the main transaction category of the bank branch and the risk control model.

[0013] In the second aspect of the embodiments of the present invention, a customer transaction risk control system for a bank branch is proposed, including:

[0014] A transaction classification module, configured to obtain transaction data of a bank branch, classify the transactions according to the transaction data, and obtain multiple transaction categories;

[0015] A current moment processing module, configured to number the time within a set period at the current moment. For each number, determine the transaction category vector corresponding to the current moment according to the transaction data of the bank branch at the time corresponding to the number;

[0016] A historical moment processing module, configured to, for multiple historical moments, number the time within a set period at the historical moment. For each number, determine the transaction category vector corresponding to the historical moment according to the transaction data of the bank branch at the time corresponding to the number;

[0017] A similar historical moment determination module, configured to determine the similar historical moment of the current moment according to the multiple transaction category vectors corresponding to the current moment and the multiple transaction category vectors corresponding to each historical moment among the multiple historical moments;

[0018] A distribution module, configured to determine the main transaction category of the bank branch and the risk control model corresponding to the main transaction category according to the transaction data after the similar historical moment, and distribute the main transaction category and the risk control model to the bank branch;

[0019] A risk control module, configured to, when a customer conducts a transaction at a bank branch, perform risk control on the current transaction of the customer according to the main transaction category of the bank branch and the risk control model.

[0020] In the third aspect of the embodiments of the present invention, a computer device is proposed, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the customer transaction risk control method for a bank branch is implemented.

[0021] In the fourth aspect of the embodiments of the present invention, a computer-readable storage medium is proposed. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the customer transaction risk control method for a bank branch is implemented.

[0022] In the fifth aspect of the embodiments of the present invention, a computer program product is proposed. The computer program product includes a computer program, and when the computer program is executed by a processor, the customer transaction risk control method for a bank branch is implemented.

[0023] The customer transaction risk control method and system for bank branches proposed by the present invention enable bank branches to control the risks of customer transactions, without uploading each transaction to the server for risk control, shortening the waiting time of customers, reducing the pressure on server resources, improving transaction processing efficiency, enhancing the customer experience, and preventing the occurrence of transaction risks. Description of the Drawings

[0024] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0025] Figure 1 It is a schematic flowchart of the customer transaction risk control method for a bank branch according to an embodiment of the present invention.

[0026] Figure 2 It is a schematic flowchart of the specific process for determining the transaction category vector corresponding to the current moment and the serial number.

[0027] Figure 3 It is a schematic flowchart of the specific process for determining the transaction category vector corresponding to the historical moment and the serial number.

[0028] Figure 4 It is a schematic flowchart of the specific process for determining the similar historical moment at the current moment.

[0029] Figure 5 It is a schematic flowchart of the specific process for distributing the main transaction categories and the risk control model to the bank branch.

[0030] Figure 6 It is a schematic flowchart of the specific process for controlling the risk of the current transaction of a customer according to the main transaction categories and the risk control model of the bank branch.

[0031] Figure 7 It is a schematic diagram of the architecture of the customer transaction risk control system for a bank branch according to an embodiment of the present invention.

[0032] Figure 8 It is a schematic diagram of the structure of a computer device according to an embodiment of the present invention. Detailed Embodiments

[0033] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and implement the present invention, and do not limit the scope of the present invention in any way. On the contrary, these embodiments are provided to make the present disclosure more thorough and complete, and to fully convey the scope of the present disclosure to those skilled in the art.

[0034] Those skilled in the art know that the embodiments of the present invention can be implemented as a system, device, equipment, method, or computer program product. Therefore, the present disclosure can be specifically implemented in the following forms: completely hardware, completely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0035] According to an embodiment of the present invention, a method and system for controlling customer transaction risks in a bank branch are proposed, which relate to the technical field of transaction data processing.

[0036] The principles and spirit of the present invention will be elaborated in detail below with reference to several representative embodiments of the present invention.

[0037] Figure 1 It is a schematic flowchart of a method for controlling customer transaction risks in a bank branch according to an embodiment of the present invention. As Figure 1 shown, the method includes:

[0038] S1. Obtain the transaction data of the bank branch, classify the transactions according to the transaction data, and obtain multiple transaction categories;

[0039] S2. Number the time within a set period at the current moment. For each number, determine the transaction category vector corresponding to the current moment at the time corresponding to the number according to the transaction data of the bank branch at that time;

[0040] S3. For multiple historical moments, number the time within a set period at that historical moment. For each number, determine the transaction category vector corresponding to that historical moment at the time corresponding to the number according to the transaction data of the bank branch at that time;

[0041] S4. Determine the similar historical moment of the current moment according to the multiple transaction category vectors corresponding to the current moment and the multiple transaction category vectors corresponding to each historical moment among the multiple historical moments;

[0042] S5. Determine the main transaction category of the bank branch and the risk control model corresponding to the main transaction category according to the transaction data after the similar historical moment, and send the main transaction category and the risk control model to the bank branch;

[0043] S6. When a customer conducts a transaction at a bank branch, risk control is performed on the customer's current transaction according to the main transaction categories and risk control models of the bank branch.

[0044] For a clearer explanation of the above customer transaction risk control method for bank branches, the following will provide a detailed description in combination with each step.

[0045] In S1, transaction data of the bank branch is obtained, and the transactions are classified according to the transaction data to obtain multiple transaction categories.

[0046] Among them, transactions can include: withdrawals, transfers, and other transactions.

[0047] First, determine the distance function corresponding to the transaction. Specifically, for each transaction, determine the number of customers belonging to each customer category in the customer set corresponding to the transaction; for each customer category, determine the distance function of the transaction with respect to the customer category, where the independent variable of the distance function is two transactions, and the corresponding function value is the absolute value of the difference in the number of customers belonging to the customer category in the customer sets corresponding to the two transactions; determine the distance function corresponding to the transaction as the square root of the weighted sum of squares of the distance functions of the transaction with respect to each customer category.

[0048] After obtaining the distance function corresponding to the determined transaction, the transaction can be classified. In one embodiment, according to the distance function corresponding to the transaction, a clustering algorithm is selected to perform clustering analysis on all transactions to obtain multiple transaction subsets, and each transaction subset corresponds to a transaction category.

[0049] To obtain more accurate classification results, for each transaction subset obtained above, determine the main risk category of each transaction in the transaction subset; select the main risk category with the largest number of transactions corresponding to the transaction subset from the main risk categories of all transactions in the transaction subset, and determine the main risk category of the transaction subset as the main risk category; for the transaction subset, determine whether the following condition a is satisfied: whether the ratio of the number of transactions corresponding to the main risk category in all transactions of the transaction subset to the number of transactions in the transaction subset is greater than a set threshold. If not satisfied (i.e., the ratio is less than or equal to the set threshold), continue to perform clustering analysis on the transaction subset until each newly generated transaction subset satisfies the above condition a.

[0050] In S2, refer to Figure 2 , number the time within the set period at the current moment. For each number, the specific process of determining the transaction category vector corresponding to the current moment for the number according to the transaction data of the bank branch at the time corresponding to the number is as follows:

[0051] S201. For each number, determine the transaction category to which each transaction in the transaction data of the bank branch at the time corresponding to the number belongs.

[0052] S202. Determine the transaction category vector corresponding to the number at the current moment; where each component of the transaction category vector corresponds one-to-one to each transaction category, and the value of each component is equal to the number of transactions in the transaction data of the bank branch at the time corresponding to the number that belong to the transaction category corresponding to the component; where the transaction categories corresponding to the same component in the transaction category vectors corresponding to each number at the current moment are the same.

[0053] In S3, refer to Figure 3 , for multiple historical moments, number the time within the set period at that historical moment. For each number, the specific process of determining the transaction category vector corresponding to the number at that historical moment according to the transaction data of the bank branch at the time corresponding to the number is as follows:

[0054] S301. For each number, determine the transaction category to which each transaction in the transaction data of the bank branch at the time corresponding to the number belongs.

[0055] S302. Determine the transaction category vector corresponding to the number at that historical moment; where each component of the transaction category vector corresponds one-to-one to each transaction category, and the value of each component is equal to the number of transactions in the transaction data of the bank branch at the time corresponding to the number that belong to the transaction category corresponding to the component; where the transaction categories corresponding to the same component in the transaction category vectors corresponding to each number at that historical moment are the same as those in the transaction category vectors corresponding to each number at the current moment.

[0056] Combining S2 and S3, an explanation of how to determine the transaction category vector is as follows:

[0057] Number the time within the set period at the current moment (or historical moment), and the time is numbered in sequence from near to far from the current moment (or historical moment) (such as 1, 2, 3, 4, 5, 6).

[0058] For example, the current moment is January 15th, and the set period is 5 days.

[0059] The numbers corresponding to the current moment of January 15th are:

[0060] January 14th, number 1;

[0061] January 13th, number 2;

[0062] January 12th, number 3;

[0063] January 11th, number 4;

[0064] On January 10th, number 5;

[0065] Multiple historical moments are January 14th, 13th, 12th, etc.

[0066] The number corresponding to the historical moment of January 14th is:

[0067] On January 13th, number 1;

[0068] On January 12th, number 2;

[0069] On January 11th, number 3;

[0070] On January 10th, number 4;

[0071] On January 9th, number 5;

[0072] And so on, the numbers corresponding to historical moments such as January 13th and 12th can be obtained. Each number corresponds to a transaction category vector.

[0073] In S4, refer to Figure 4 , according to the multiple transaction category vectors corresponding to the current moment and the multiple transaction category vectors corresponding to each historical moment among the multiple historical moments, the specific process for determining the similar historical moment of the current moment is as follows:

[0074] S401, according to the multiple transaction category vectors corresponding to the current moment and the multiple transaction category vectors corresponding to each historical moment among the multiple historical moments, determine the partial order of the historical moments, where the partial order is used to determine whether the first historical moment is closer to the second historical moment among any two historical moments in the multiple historical moments;

[0075] S402, based on the partial order of the historical moments, determine the maximal historical moments among the multiple historical moments, where the maximal historical moment is the maximal element of the partial order;

[0076] It should be noted that the maximal element of the partial order is an element in the set corresponding to the partial order for which there is no other element closer to it.

[0077] S403, determine the maximal historical moment among the multiple historical moments as the similar historical moment of the current moment.

[0078] In S401, the detailed process of determining the partial order of the historical moments according to the multiple transaction category vectors corresponding to the current moment and the multiple transaction category vectors corresponding to each historical moment among the multiple historical moments is as follows:

[0079] S4011. For each number, determine the distance between the transaction category vector corresponding to each historical moment in the multiple historical moments and the transaction category vector corresponding to the current moment for that number, and use this distance as the distance corresponding to that historical moment for that number.

[0080] S4012. Determine the partial order of the historical moments. Among them, for any two historical moments in the multiple historical moments, if for each number, the distance corresponding to the first historical moment of the two historical moments is less than or equal to the distance corresponding to the second historical moment of the two historical moments for that number, then determine that the first historical moment is closer to the second historical moment.

[0081] In S402, according to the partial order of the historical moments, the specific method for determining the maximal historical moments in the multiple historical moments is as follows:

[0082] S4021. Initialize by setting the maximal flag corresponding to each historical moment in the multiple historical moments to possible, and setting the comparison flag corresponding to each historical moment to yes.

[0083] S4022. For each historical moment in the multiple historical moments in turn, select from all other historical moments in the multiple historical moments except this historical moment the multiple historical moments whose corresponding comparison flags are yes, then set the historical moments to be compared corresponding to this historical moment to the selected multiple historical moments, and then execute the following step S4023; if the maximal flag corresponding to this historical moment is not possible, continue to execute step S4022 for the next historical moment.

[0084] S4023. Select each historical moment to be compared corresponding to this historical moment in turn, and confirm whether this historical moment to be compared is closer to this historical moment; if this historical moment to be compared is closer to this historical moment, set the maximal flag corresponding to this historical moment to no, and then execute the above step S4022; if this historical moment is closer to this historical moment to be compared, set the maximal flag corresponding to this historical moment to be compared to no, and add this historical moment to be compared to the set of sub-historical moments corresponding to this historical moment; otherwise, the maximal flag corresponding to this historical moment and the maximal flag corresponding to this historical moment to be compared remain unchanged.

[0085] S4024. If it is confirmed that all historical moments to be compared corresponding to this historical moment are not closer to this historical moment (that is, after comparing this historical moment with each corresponding historical moment to be compared in turn, the maximal flag corresponding to this historical moment is still possible), then determine this historical moment as the maximal historical moment in the multiple historical moments, and update the comparison flag of each historical moment in the set of sub-historical moments of this maximal historical moment to no.

[0086] At S4025, steps S4022 are then continued for the next historical moment until steps S4022, S4023, and S4024 are executed for all historical moments among multiple historical moments.

[0087] It should be noted that the above partial order can determine the similarity between the current moment and each historical moment. There may be multiple maximal elements of the partial order. Using the maximal elements of this partial order to determine the similar historical moments of the current moment can integrate the characteristics of each similar historical moment and avoid information loss caused by selecting a single similar historical moment.

[0088] In S5, refer to Figure 5 , and according to the transaction data after the similar historical moment, determine the main transaction category of the bank branch and the risk control model corresponding to the main transaction category. The specific process of sending the main transaction category and the risk control model to the bank branch is as follows:

[0089] S501, From the transaction data after the similar historical moment, select the transaction category with the largest trading volume, and use this transaction category as the first transaction category after the current moment;

[0090] S502, Determine the controllable transaction categories of the first transaction category, and use the first transaction category and the controllable transaction categories as the main transaction categories after the current moment;

[0091] S503, According to the transaction data of the first transaction category, determine the risk control model corresponding to the main transaction category.

[0092] In S502, the detailed process of determining the controllable transaction categories of the first transaction category and using the first transaction category and the controllable transaction categories as the main transaction categories after the current moment is as follows:

[0093] S5021, For each transaction category, obtain the transaction data of the transaction category, and determine the risk type of the transaction category and the risk probability corresponding to each risk type;

[0094] S5022, For each transaction category, select from the risk types of the transaction category the risk types whose corresponding risk probabilities are greater than the set threshold as the main risk types corresponding to the transaction category;

[0095] S5023, For each transaction category, if the main risk types corresponding to the transaction category are all included in the main risk types of the first transaction category, determine the transaction category as the controllable transaction category of the first transaction category.

[0096] In S5021, the transaction data of this transaction category refers to the transaction data of the bank branch regarding this transaction category. The risk types involved in the transaction data are determined as the risk types of this transaction category. To obtain a more accurate risk probability, the steps to determine the risk probability corresponding to each risk type are as follows:

[0097] Determine the distance function between bank branches;

[0098] For each customer of bank branch A, obtain the transaction data of this customer, and determine the ratio of the number of transactions involving each risk type in this transaction data to the corresponding number of transactions in this transaction data. Determine this ratio as the risk probability of this customer regarding each risk type (step s);

[0099] For each risk type, consider the risk probabilities of all customers of bank branch A regarding this risk type as multiple samples of the same probability distribution. Based on these multiple samples, the variance corresponding to this risk type can be determined;

[0100] Set the maximum value of the variances corresponding to all risk types as σ;

[0101] Set the customer number threshold according to this variance σ

[0102] where ε is the acceptable risk probability error threshold, and P is the probability that the acceptable risk probability error is greater than ε;

[0103] Determine the magnitude relationship between the number of customers of bank branch A and this customer number threshold. If the number of customers of bank branch A is greater than or equal to this customer number threshold, then for each risk type, calculate the mean value of the risk probabilities of all customers of bank branch A regarding this risk type, and determine this mean value as the risk probability corresponding to this risk type;

[0104] If the number of customers of bank branch A is less than this customer number threshold, determine the distance between bank branch A and other bank branches based on the distance function between bank branches. Then set a distance threshold such that the sum of the number of customers of multiple similar bank branches whose distance from bank branch A is less than the distance threshold and bank branch A is greater than or equal to this customer number threshold; for each risk type, calculate the mean value of the risk probabilities of all customers of multiple similar bank branches and bank branch A regarding this risk type, and determine this mean value as the risk probability corresponding to this risk type. Among them, the calculation method of the risk probabilities of the customers of these multiple similar bank branches regarding each risk type is the same as the calculation method of the risk probabilities of each customer of bank branch A regarding each risk type. For details, please refer to step s above.

[0105] According to the law of large numbers, the above method can calculate a more accurate risk probability.

[0106] Among them, the distance function for determining between bank branches can be: for each bank branch, obtain the transaction data of this bank branch, and determine the number of transactions corresponding to each transaction category in this transaction data; determine the distance function corresponding to the bank branch, where the independent variable of this distance function is two bank branches, and the corresponding function value is the square root of the weighted sum of squares of the differences in the number of transactions belonging to each transaction category in the transaction data of these two bank branches.

[0107] It should be noted that the main risk type corresponding to each transaction category is the risk type that needs to be controlled in real time for this transaction category. If for a certain transaction category, all the main risk types corresponding to this transaction category are included in the main risk types of this first transaction category, it means that all the risk types that need to be controlled in real time for this transaction category are included in the risk types that need to be controlled in real time for this first transaction category.

[0108] In S503, according to the transaction data of this first transaction category, determine the risk control model corresponding to this main transaction category, including:

[0109] S5031, train a risk prediction model according to the transaction data of this first transaction category, where the output of this risk prediction model is the risk probability;

[0110] S5032, set the proportion of risk data in the transaction data of this first transaction category as the risk threshold; where when the risk probability output by the risk prediction model is greater than this risk threshold, it is determined that the transaction is risky and the corresponding response method is initiated, otherwise it is determined to be risk-free.

[0111] The response method can include the bank system rejecting the customer's transaction, manually reviewing the customer's transaction, or authenticating the customer's identity (such as, live body recognition).

[0112] In S6, refer to Figure 6 , when a customer conducts a transaction at a bank branch, the specific process of controlling the risk of the customer's current transaction according to the main transaction category and risk control model of this bank branch is as follows:

[0113] S601, determine whether the transaction category to which the current transaction belongs is among the main transaction categories of this bank branch;

[0114] S602, if it is, use this risk control model to control the risk of this current transaction; otherwise, send the transaction data of this current transaction to the bank back-end server to obtain the risk control result.

[0115] Based on the above method, the bank branch can control the risks of customer transactions without uploading each transaction to the server for risk control, shortening the waiting time of customers and reducing the pressure on server resources. Through the analysis of transaction data, the main transaction categories of the bank branch can be obtained, and the main transaction categories include the first transaction category and the controllable transaction category; the risks to be controlled in the controllable transaction category are weaker than those in the first transaction category; that is, the risk prediction model of the first transaction category can also predict the risks of the controllable transaction category.

[0116] In the actual application scenario, the bank server is set with risk prediction models for all transaction categories. In this regard, the present invention analyzes the transaction data, and distributes the main transaction categories and risk predictions to the bank branch, enabling the bank branch to control the risks of the main transaction categories, effectively reducing the waiting time of customers, improving the customer experience and preventing the occurrence of transaction risks.

[0117] It should be noted that although the operations of the method of the present invention are described in a specific order in the above embodiments and the accompanying drawings, this does not require or imply that these operations must be performed in this specific order, or that all the operations shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step for execution, and / or one step may be decomposed into multiple steps for execution.

[0118] After introducing the method of the exemplary embodiment of the present invention, next, reference is made to Figure 7 to introduce the customer transaction risk control system of the bank branch of the exemplary embodiment of the present invention.

[0119] The implementation of the customer transaction risk control system of the bank branch can refer to the implementation of the above method, and the repeated parts will not be elaborated. The terms "module" or "unit" used hereinafter may be a combination of software and / or hardware for implementing a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0120] Based on the same inventive concept, the present invention also proposes a customer transaction risk control system of a bank branch, as Figure 7 shown, the system includes:

[0121] A transaction classification module 710, configured to obtain transaction data of a bank branch, classify the transactions according to the transaction data, and obtain multiple transaction categories;

[0122] A current moment processing module 720, configured to number the time within a set period at the current moment, and for each number, determine the transaction category vector corresponding to the current moment according to the transaction data of the bank branch at the time corresponding to the number;

[0123] A historical moment processing module 730, configured to number the time within a set period at each of multiple historical moments, and for each number, determine a transaction category vector corresponding to the historical moment for the number according to the transaction data of the bank branch at the time corresponding to the number;

[0124] A similar historical moment determination module 740, configured to determine a similar historical moment of the current moment according to the multiple transaction category vectors corresponding to the current moment and the multiple transaction category vectors corresponding to each of the multiple historical moments;

[0125] A sending module 750, configured to determine a main transaction category of the bank branch and a risk control model corresponding to the main transaction category according to the transaction data after the similar historical moment, and send the main transaction category and the risk control model to the bank branch;

[0126] A risk control module 760, configured to perform risk control on the current transaction of a customer according to the main transaction category of the bank branch and the risk control model when the customer conducts a transaction at the bank branch.

[0127] In an embodiment, the current moment processing module is specifically configured to:

[0128] For each number, determine the transaction category to which each transaction in the transaction data of the bank branch at the time corresponding to the number belongs;

[0129] Determine the transaction category vector corresponding to the current moment for the number; wherein, each component of the transaction category vector corresponds to each transaction category one by one, and the value of each component is equal to the number of transactions in the transaction data of the bank branch at the time corresponding to the number that belong to the transaction category corresponding to the component; wherein, the transaction category vectors corresponding to the current moment for each number have the same transaction category corresponding to the same component.

[0130] In an embodiment, the historical moment processing module is specifically configured to:

[0131] For each number, determine the transaction category to which each transaction in the transaction data of the bank branch at the time corresponding to the number belongs;

[0132] Determine the transaction category vector corresponding to the historical moment for the number; wherein, each component of the transaction category vector corresponds to each transaction category one by one, and the value of each component is equal to the number of transactions in the transaction data of the bank branch at the time corresponding to the number that belong to the transaction category corresponding to the component; wherein, the transaction category vectors corresponding to the historical moment for each number and the transaction category vectors corresponding to the current moment for each number have the same transaction category corresponding to the same component.

[0133] In one embodiment, the similar historical moment determining module is specifically configured to:

[0134] Determine the partial order of historical moments according to the multiple transaction category vectors corresponding to the current moment and the multiple transaction category vectors corresponding to each historical moment among the multiple historical moments, where the partial order is used to determine whether a first historical moment is closer to a second historical moment among any two historical moments of the multiple historical moments;

[0135] Determine the maximum historical moment among the multiple historical moments according to the partial order of historical moments, where the maximum historical moment is the maximum element of the partial order;

[0136] Determine the maximum historical moment among the multiple historical moments as the similar historical moment of the current moment.

[0137] In one embodiment, the similar historical moment determining module is further specifically configured to:

[0138] For each number, determine the distance between the transaction category vector corresponding to each historical moment among the multiple historical moments and the transaction category vector corresponding to the current moment for this number, and use this distance as the distance corresponding to this number for this historical moment;

[0139] Determine the partial order of historical moments, where for any two historical moments among the multiple historical moments, if for each number, the distance corresponding to this number for the first historical moment of the two historical moments is less than or equal to the distance corresponding to this number for the second historical moment of the two historical moments, then determine that the first historical moment is closer to the second historical moment.

[0140] In one embodiment, the sending module is specifically configured to:

[0141] Select the transaction category with the largest trading volume from the transaction data after the similar historical moment, and use this transaction category as the first transaction category after the current moment;

[0142] Determine the controllable transaction category of the first transaction category, and use the first transaction category and the controllable transaction category as the main transaction categories after the current moment;

[0143] Determine the risk control model corresponding to the main transaction categories according to the transaction data of the first transaction category.

[0144] In one embodiment, the sending module is specifically configured to:

[0145] For each transaction category, obtain the transaction data of this transaction category, and determine the risk type of this transaction category and the risk probability corresponding to each risk type;

[0146] For each transaction category, select, from the risk types of that transaction category, the risk types for which the corresponding risk probabilities are greater than a set threshold as the main risk types corresponding to that transaction category;

[0147] For each transaction category, if the main risk types corresponding to that transaction category are all included in the main risk types of the first transaction category, determine that transaction category as a controllable transaction category of the first transaction category.

[0148] In an embodiment, the risk control module is specifically configured to:

[0149] Determine whether the transaction category to which the current transaction belongs is among the main transaction categories of the bank branch;

[0150] If so, perform risk control on the current transaction using the risk control model; otherwise, send the transaction data of the current transaction to the bank's back-end server to obtain the risk control result.

[0151] It should be noted that although several modules of the customer transaction risk control system of the bank branch are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, according to the embodiments of the present invention, the features and functions of the two or more modules described above can be embodied in one module. Conversely, the features and functions of one module described above can be further divided and embodied by multiple modules.

[0152] Based on the foregoing inventive concept, as Figure 8 shown, the present invention also proposes a computer device 800, including a memory 810, a processor 820, and a computer program 830 stored in the memory 810 and executable on the processor 820. When the processor 820 executes the computer program 830, the foregoing customer transaction risk control method of the bank branch is implemented.

[0153] Based on the foregoing inventive concept, the present invention proposes a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the foregoing customer transaction risk control method of the bank branch is implemented.

[0154] Based on the foregoing inventive concept, the present invention proposes a computer program product including a computer program, and when the computer program is executed by a processor, the customer transaction risk control method of the bank branch is implemented.

[0155] The customer transaction risk control method and system for bank branches proposed by the present invention can enable bank branches to control the risks of customer transactions, without uploading each transaction to the server for risk control, shortening the waiting time of customers, reducing the pressure on server resources, improving the transaction processing efficiency, enhancing the customer experience, and preventing the occurrence of transaction risks.

[0156] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can 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.

[0157] The present invention is described with reference to the flowcharts and / or block diagrams of methods and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows 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 the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the specified functions in Figure 1 one or more of the processes Figure 1 or a plurality of processes and / or blocks

[0158] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means implements the specified functions in Figure 1 one or more of the processes Figure 1 or a plurality of processes and / or blocks

[0159] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the specified functions in Figure 1 one or more of the processes Figure 1 or a plurality of processes and / or blocks

[0160] Finally, it should be noted that the above-described embodiments are only specific embodiments of the present invention, used to illustrate the technical solutions of the present invention, rather than limiting it. The protection scope of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that any person skilled in the art within the technical scope disclosed by the present invention can still modify the technical solutions recorded in the foregoing embodiments, or can easily think of changes, or make equivalent replacements for some of the technical features; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A method for controlling customer transaction risks at bank branches, characterized in that, it includes: Obtain the transaction data of the bank branch, classify the transactions according to the transaction data, and obtain multiple transaction categories; Number the time within a set period at the current moment. For each number, determine the transaction category vector corresponding to the current moment according to the transaction data of the bank branch at the time corresponding to the number; For multiple historical moments, number the time within a set period at the historical moment. For each number, determine the transaction category vector corresponding to the historical moment according to the transaction data of the bank branch at the time corresponding to the number; Determine the similar historical moment of the current moment according to the multiple transaction category vectors corresponding to the current moment and the multiple transaction category vectors corresponding to each historical moment among the multiple historical moments; Determine the main transaction category of the bank branch and the risk control model corresponding to the main transaction category according to the transaction data after the similar historical moment, and send the main transaction category and the risk control model to the bank branch; When a customer conducts a transaction at the bank branch, control the risk of the customer's current transaction according to the main transaction category of the bank branch and the risk control model; Among them, for multiple historical moments, number the time within a set period at the historical moment. For each number, determine the transaction category vector corresponding to the historical moment according to the transaction data of the bank branch at the time corresponding to the number, including: For each number, determine the transaction category to which each transaction in the transaction data of the bank branch at the time corresponding to the number belongs; Determine the transaction category vector corresponding to the historical moment; among them, each component of the transaction category vector corresponds one by one to each transaction category, and the value of each component is equal to the number of transactions in the transaction data of the bank branch at the time corresponding to the number that belong to the transaction category corresponding to the component; among them, the transaction category vectors corresponding to each number of the historical moment and the transaction category vectors corresponding to each number of the current moment are the same in the transaction category corresponding to the same component; Among them, determining the similar historical moment of the current moment according to the multiple transaction category vectors corresponding to the current moment and the multiple transaction category vectors corresponding to each historical moment among the multiple historical moments includes: Determine the partial order of the historical moments according to the multiple transaction category vectors corresponding to the current moment and the multiple transaction category vectors corresponding to each historical moment among the multiple historical moments, where the partial order is used to determine whether the first historical moment is closer to the second historical moment among any two historical moments in the multiple historical moments; According to the partial order of the historical moments, determine the maximal historical moments among the multiple historical moments, where the maximal historical moment is the maximal element of the partial order; Determine the maximal historical moment among the multiple historical moments as the similar historical moment of the current moment; Among them, determining the main transaction category of the bank branch and the risk control model corresponding to the main transaction category according to the transaction data after the similar historical moment, and sending the main transaction category and the risk control model to the bank branch includes: Select the transaction category with the largest trading volume from the transaction data after a similar historical moment, and use this transaction category as the first transaction category after the current moment; Determine the controllable transaction categories of the first transaction category, and use the first transaction category and the controllable transaction categories as the main transaction categories after the current moment; Determine the risk control model corresponding to the main transaction category according to the transaction data of the first transaction category; Among them, determining the controllable transaction categories of the first transaction category and using the first transaction category and the controllable transaction categories as the main transaction categories after the current moment includes: For each transaction category, obtain the transaction data of this transaction category, and determine the risk type of this transaction category and the risk probability corresponding to each risk type; For each transaction category, select from the risk types of this transaction category the risk types whose corresponding risk probabilities are greater than the set threshold as the main risk types corresponding to this transaction category; For each transaction category, if the main risk types corresponding to this transaction category are all included in the main risk types of the first transaction category, determine this transaction category as the controllable transaction category of the first transaction category.

2. The method according to claim 1, characterized in that, Number the time within the set period at the current moment. For each number, determine the transaction category vector corresponding to the current moment at the time corresponding to this number according to the transaction data of the bank branch at the time corresponding to this number, including: For each number, determine the transaction categories to which each transaction in the transaction data of the bank branch at the time corresponding to this number belongs; Determine the transaction category vector corresponding to the current moment at the time corresponding to this number; among them, each component of the transaction category vector corresponds one-to-one with each transaction category, and the value of each component is equal to the number of transactions belonging to the transaction category corresponding to this component in the transaction data of the bank branch at the time corresponding to this number; among them, the transaction categories corresponding to the same component of the transaction category vectors corresponding to the current moment at each number are the same.

3. The method according to claim 1, characterized in that, Determine the partial order of historical moments according to the multiple transaction category vectors corresponding to the current moment and the multiple transaction category vectors corresponding to each historical moment in multiple historical moments, including: For each number, determine the distance between the transaction category vector corresponding to this number of each historical moment in the multiple historical moments and the transaction category vector corresponding to this number of the current moment, and use this distance as the distance corresponding to this number of this historical moment; Determine the partial order of historical moments. Among them, for any two historical moments in the multiple historical moments, if for each number, the distance corresponding to this number of the first historical moment of the two historical moments is less than or equal to the distance corresponding to this number of the second historical moment of the two historical moments, then determine that the first historical moment is closer to the second historical moment.

4. The method according to claim 1, characterized in that, When a customer conducts a transaction at a bank branch, perform risk control on the customer's current transaction according to the main transaction category and risk control model of this bank branch, including: Determine whether the transaction category to which the current transaction belongs is among the main transaction categories of the bank branch; If so, use the risk control model to perform risk control on the current transaction; otherwise, send the transaction data of the current transaction to the bank's back-end server to obtain the risk control result.

5. A customer transaction risk control system for a bank branch, Characterized in that, It includes: A transaction classification module, configured to obtain transaction data of a bank branch, classify the transactions according to the transaction data, and obtain multiple transaction categories; A current moment processing module, configured to number the time within a set period at the current moment, and for each number, determine the transaction category vector corresponding to the current moment according to the transaction data of the bank branch at the time corresponding to the number; A historical moment processing module, configured to, for multiple historical moments, number the time within a set period at the historical moment, and for each number, determine the transaction category vector corresponding to the historical moment according to the transaction data of the bank branch at the time corresponding to the number; A similar historical moment determination module, configured to determine the similar historical moment of the current moment according to the multiple transaction category vectors corresponding to the current moment and the multiple transaction category vectors corresponding to each historical moment among the multiple historical moments; A sending module, configured to determine the main transaction category of the bank branch and the risk control model corresponding to the main transaction category according to the transaction data after the similar historical moment, and send the main transaction category and the risk control model to the bank branch; A risk control module, configured to, when a customer conducts a transaction at the bank branch, perform risk control on the current transaction of the customer according to the main transaction category of the bank branch and the risk control model; Wherein, the historical moment processing module is specifically configured to: For each number, determine the transaction category to which each transaction in the transaction data of the bank branch at the time corresponding to the number belongs; Determine the transaction category vector corresponding to the historical moment; wherein, each component of the transaction category vector corresponds to each transaction category one by one, and the value of each component is equal to the number of transactions belonging to the transaction category corresponding to the component in the transaction data of the bank branch at the time corresponding to the number; wherein, the transaction category vectors corresponding to each number of the historical moment and the transaction category vectors corresponding to each number of the current moment have the same transaction category corresponding to the same component; Wherein, the similar historical moment determination module is specifically configured to: Determine the partial order of the historical moments according to the multiple transaction category vectors corresponding to the current moment and the multiple transaction category vectors corresponding to each historical moment among the multiple historical moments, wherein the partial order is used to determine whether the first historical moment is closer to the second historical moment among any two historical moments in the multiple historical moments; According to the partial order of the historical moments, determine the maximum historical moment among the multiple historical moments, wherein the maximum historical moment is the maximum element of the partial order; Determine the maximum historical moment among the multiple historical moments as the similar historical moment of the current moment; Wherein, the sending module is specifically configured to: Select the transaction category with the largest trading volume from the transaction data after a similar historical moment, and use this transaction category as the first transaction category after the current moment; Determine the controllable transaction categories of this first transaction category, and use this first transaction category and the controllable transaction categories as the main transaction categories after the current moment; Determine the risk control model corresponding to this main transaction category according to the transaction data of this first transaction category; Among them, the sending module is specifically used for: For each transaction category, obtain the transaction data of this transaction category, and determine the risk type of this transaction category and the risk probability corresponding to each risk type; For each transaction category, select, from the risk types of this transaction category, the risk types whose corresponding risk probabilities are greater than the set threshold as the main risk types corresponding to this transaction category; For each transaction category, if the main risk types corresponding to this transaction category are all included in the main risk types of this first transaction category, determine this transaction category as the controllable transaction category of this first transaction category.

6. The system according to claim 5, characterized in that, The current moment processing module is specifically used for: For each number, determine the transaction categories to which each transaction belongs in the transaction data of the bank branch at the time corresponding to this number; Determine the transaction category vector corresponding to the current moment for this number; wherein, each component of this transaction category vector corresponds one-to-one with each transaction category, and the value of each component is equal to the number of transactions belonging to the transaction category corresponding to this component in the transaction data of the bank branch at the time corresponding to this number; wherein, the transaction categories corresponding to the same component of the transaction category vectors corresponding to the current moment for each number are the same.

7. The system according to claim 5, characterized in that, The similar historical moment determination module is specifically used for: For each number, determine the distance between the transaction category vector corresponding to this number at each historical moment among the multiple historical moments and the transaction category vector corresponding to the current moment for this number, and use this distance as the distance corresponding to this number at this historical moment; Determine the partial order of historical moments, wherein, for any two historical moments among the multiple historical moments, if for each number, the distance corresponding to this number at the first historical moment of the two historical moments is less than or equal to the distance corresponding to this number at the second historical moment of the two historical moments, then determine that the first historical moment is closer to the second historical moment.

8. The system according to claim 5, characterized in that, The risk control module is specifically used for: Judge whether the transaction category to which the current transaction belongs is among the main transaction categories of this bank branch; If so, perform risk control on this current transaction using this risk control model; otherwise, send the transaction data of this current transaction to the bank background server to obtain the risk control result.

9. A computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method according to any one of claims 1 to 4.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

11. A computer program product, characterized in that the computer program product includes a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images