Account early warning method, device, electronic device and storage medium

Abstract

The present invention provides an account early warning method, device, electronic device and storage medium, which can be applied to the field of artificial intelligence or the financial field. In the transfer process of the target account, it can determine whether it hits the strong rule of erroneous transfer according to the first transfer information of the account. If it hits, the second transfer information of the account is predicted to be wrong, so as to complete the early warning of the account. Since the amount of information of the second transfer information is higher than the amount of information of the first transfer information, after a small amount of transfer information hits the strong rule, a large amount of transfer information is predicted to be wrong, which can accurately and quickly identify the erroneous transfer of the account and improve the security of the transfer.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and more specifically, to an account early warning method, device, electronic device and storage medium. Background Art

[0002] The bank's transfer system uses the BGL account as the transition system. However, due to bank requirements, the BGL account does not have a limit setting. As a result, there are cases where bank staff mistakenly fill in the bank card number as the transfer amount, thereby transferring a large amount of cash to the customer, causing disputes, which has a very bad impact on the bank. Summary of the invention

[0003] In view of this, in order to solve the above problems, the present invention provides an account early warning method, device, electronic device and storage medium, and the technical solution is as follows:

[0004] An account early warning method, the method comprising:

[0005] Obtain the first transfer information of the target account;

[0006] If it is determined that the first transfer information hits the configured strong rule, obtaining second transfer information of the target account, where the amount of information of the second transfer information is greater than the amount of information of the first transfer information;

[0007] An error prediction is performed on the second transfer information, and a processing operation is performed on the target account according to the prediction result.

[0008] Preferably, the configuration method of the strong rule includes:

[0009] Extracting feature parameters from the first information of the transferor in the transfer dispute event, wherein the feature parameters include amount features and account features;

[0010] Analyze the amount characteristics and the account characteristics respectively, and determine the transfer amount rules and transfer account rules for the transfer dispute case;

[0011] Accordingly, determining that the first transfer information hits a configured strong rule includes:

[0012] determining whether the transfer amount in the first transfer information matches the transfer amount rule; or,

[0013] It is determined that the transfer account in the first transfer information matches the transfer account rule.

[0014] Preferably, the error prediction of the second transfer information includes:

[0015] Retrieving a prediction model, wherein the prediction model is obtained by training network parameters of a basic network model by using the second information of the transfer parties in the transfer dispute event as training samples and aiming to make the error prediction results of the training samples close to their error labeling results;

[0016] The second transfer information is input into the prediction model to obtain an error prediction result of the second transfer information through the prediction model.

[0017] Preferably, the basic network model includes an input layer, a hidden layer and an output layer, and the initial network parameters of the hidden layer are calculated by a genetic algorithm.

[0018] Preferably, executing the processing operation for the target account according to the prediction result includes:

[0019] In the event of an erroneous transfer to the target account, an alarm recheck is performed.

[0020] An account early warning device, the device comprising:

[0021] A strong rule module, configured to obtain first transfer information of a target account; if it is determined that the first transfer information hits a configured strong rule, obtain second transfer information of the target account, wherein the amount of information of the second transfer information is greater than the amount of information of the first transfer information;

[0022] The prediction module is used to predict errors of the second transfer information and execute a processing operation for the target account according to the prediction result.

[0023] Preferably, the configuration method of the strong rule includes:

[0024] Extracting characteristic parameters from the first information of the transferor in the transfer dispute event, the characteristic parameters including amount characteristics and account characteristics; analyzing the amount characteristics and the account characteristics respectively to determine the transfer amount rule and the transfer account rule of the transfer dispute case;

[0025] Accordingly, the strong rule module for determining whether the first transfer information hits a configured strong rule is specifically configured to:

[0026] determining whether the transfer amount in the first transfer information matches the transfer amount rule; or,

[0027] It is determined that the transfer account in the first transfer information matches the transfer account rule.

[0028] Preferably, the prediction module for predicting errors of the second transfer information is specifically used for:

[0029] Retrieve a prediction model, wherein the prediction model is obtained by training network parameters of a basic network model using the second information of the transfer parties in a transfer dispute event as training samples, with the goal of making the error prediction results of the training samples close to their error labeling results; input the second transfer information into the prediction model to obtain the error prediction result of the second transfer information through the prediction model.

[0030] An electronic device comprises: at least one memory and at least one processor; the memory stores a program, the processor calls the program stored in the memory, and the program is used to implement the account early warning method.

[0031] A storage medium stores computer executable instructions, and the computer executable instructions are used to execute the account early warning method.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] The present invention provides an account early warning method, device, electronic device and storage medium. In the transfer process of a target account, it is possible to determine whether the first transfer information of the account hits the strong rule of erroneous transfer according to the first transfer information of the account. If it hits, the second transfer information of the account is predicted to be wrong, so as to complete the early warning of the account. Since the amount of information of the second transfer information is higher than the amount of information of the first transfer information, after a small amount of transfer information hits the strong rule, a large amount of transfer information is predicted to be wrong, which can accurately and quickly identify the erroneous transfer of the account and improve the security of the transfer. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0035] Figure 1 A method flow chart of an account early warning method provided by an embodiment of the present invention;

[0036] Figure 2 A partial method flow chart of the account early warning method provided by an embodiment of the present invention;

[0037] Figure 3 Another method flow chart of the account early warning method provided by an embodiment of the present invention;

[0038] Figure 4 A schematic diagram of the structure of an account early warning device provided by an embodiment of the present invention;

[0039] Figure 5 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0042] The embodiment of the present invention provides an account early warning method, the method flow chart of which is as follows: Figure 1 As shown, the following steps are included:

[0043] S10, obtaining first transfer information of the target account.

[0044] In the embodiment of the present invention, the BGL account of the transferor, i.e., the target account, responds to the transfer instruction and transfers the amount under the account to the payee before executing the account warning of the present invention. First, a small amount of transfer information of the target account, i.e., the first transfer information, is obtained and a strong rule check is performed. The first transfer information may include the transfer amount of the transferor and the transfer account (i.e., the payee's account).

[0045] S20: When it is determined that the first transfer information matches the configured strong rule, obtain second transfer information of the target account, where the amount of information of the second transfer information is greater than that of the first transfer information.

[0046] In the embodiment of the present invention, by collecting case information of transfer disputes that have occurred in the past, the notary public extracts the characteristic parameters of the transferor to configure strong rules. Specifically, the configuration method of strong rules includes:

[0047] Extracting feature parameters from the first information of the transferor in the transfer dispute event, the feature parameters including amount feature and account feature; analyzing the amount feature and account feature respectively, and determining the transfer amount rule and transfer account rule of the transfer dispute case;

[0048] In the embodiment of the present invention, relevant information of the transferor, i.e., first information, is obtained from the case information of the transfer dispute event, and characteristic parameters are extracted from the first information, the characteristic parameters being the amount characteristics under the transfer amount and the account characteristics under the transfer account. Thus, the change rules of the transferor in the transfer amount and the change rules of the transfer account when an erroneous transaction occurs can be analyzed, thereby summarizing the transfer amount rule and the transfer account rule.

[0049] For example, in terms of transfer amount, a transfer amount that is too large or too small may be caused by the bank staff's misoperation. Therefore, the transfer amount rule can set an upper limit and a lower limit for the transfer amount. The values ​​of the upper limit and the lower limit of the transfer amount can be determined according to the mean distribution method, that is, the probability that the transfer amount of a transfer dispute event is between the upper limit and the lower limit of the transfer amount is less than 0.002.

[0050] In addition, in terms of transfer accounts, compare the transfer account of this transfer with multiple transfer accounts of the transferor's historical transfers. For example, take the transfer account number as an example. Compare the transfer account of this transfer with the transferor's five most recent transfer accounts. If a few digits in the transfer account number are different and most digits are the same, it may be caused by the bank clerk's error.

[0051] Accordingly, in step S20, “determining whether the first transfer information matches the configured strong rule” includes the following steps:

[0052] Determining that the transfer amount in the first transfer information matches the transfer amount rule; or,

[0053] It is determined that the transfer account in the first transfer information matches the transfer account rule.

[0054] In the embodiment of the present invention, if the transfer amount in the first transfer information is higher than the upper limit of the transfer amount, or lower than the lower limit of the transfer amount, then it is determined that the transfer amount rule is hit. If the transfer account corresponding to the transfer account in the first transfer information is different from any of the five transfer account numbers of the transferor's most recent transfer in a few digits but the same in most digits, such as one digit is different from a certain transfer account number but the other digits are the same, then it is determined that the transfer account rule is hit.

[0055] S30, predicting an error in the second transfer information, and executing a processing operation on the target account according to the prediction result.

[0056] In the embodiment of the present invention, if the first transfer information of the target account meets the rule, more detailed transfer information of the target account, i.e., the second transfer information, is read. The neural network model is used to predict whether the current transaction of the target account is an erroneous transaction. If so, the transfer is required to be checked by relevant personnel to improve the security of the transfer.

[0057] In the specific implementation process, step S30 "predicting an error in the second transfer information" can adopt the following steps, and the method flow chart is as follows Figure 2 As shown:

[0058] S301, retrieve a prediction model, wherein the prediction model is obtained by training the network parameters of a basic network model with the second information of the two parties in the transfer dispute event as training samples and with the goal of making the error prediction results of the training samples close to their error labeling results.

[0059] In the embodiment of the present invention, case information of transfer disputes that have occurred in the past is collected, and relevant information of the two parties of the transfer, that is, the second information, is obtained from the case information, and the second information includes the personal assets of the transferor, the relationship between the transferor and the payee, the relationship between the transfer account of the transferor and its previous transfer account, etc. The feature parameters of the second information are extracted to establish a feature parameter library for training the prediction model.

[0060] In the process of training the prediction model, the existing supervised learning method is used to train the basic network model, and the training samples are labeled with labels representing erroneous transactions, i.e., the erroneous labeling results. The prediction model has the ability to make the erroneous prediction results of the training samples close to their erroneous labeling results.

[0061] In addition, in the embodiment of the present invention, the basic network model includes an input layer, a hidden layer and an output layer, and the initial network parameters of the hidden layer are calculated by a genetic algorithm. The basic network model is illustrated as a BP neural network:

[0062] The embodiment of the present invention combines the advantages of BP neural network and genetic algorithm, introduces genetic algorithm in the optimization of weights and thresholds of BP neural network, and constructs GA-BP neural network model. The structure of GA-BP neural network is determined, and the structure of BP neural network is determined according to the number of network input and output, thereby determining the number of parameters that need to be optimized in genetic algorithm.

[0063] According to the Kolmogorov principle, a three-layer BP neural network is sufficient to complete any mapping from n-dimension to m-dimension, and generally only one hidden layer is needed. Therefore, in the embodiment of the present invention, the number of extracted feature data is used as the number of input layer nodes, and whether the transfer is an erroneous transfer is predicted. The number of hidden layer nodes is determined by trial and error, thereby determining the GA-BP neural network structure. The optimal individual output by the genetic algorithm is used as the initial weight and initial threshold of the hidden layer in the BP neural network to perform BP neural network training and learning.

[0064] In this regard, based on the data analysis of the feature parameter library, the GA-BP neural network model is trained, and the prediction accuracy of the model is verified using test samples. The GA-BP neural network model is used to predict whether the transfer is an erroneous transfer. This method can reduce BGL account transfer errors and improve the bank's image.

[0065] S302: Input the second transfer information into a prediction model to obtain an error prediction result of the second transfer information through the prediction model.

[0066] In the embodiment of the present invention, a prediction model is used to predict whether an erroneous transaction occurs in the second transfer information of the current transfer, and the prediction model is used to predict whether the current transfer needs to be returned for re-inspection.

[0067] In some other embodiments, in order to implement the return and re-examination of erroneous transactions, Figure 1 Based on the account warning method shown in FIG. 1 , the following steps are also included. The method flow chart is as follows: Figure 3 As shown:

[0068] S40, in case of an erroneous transfer to the target account, an alarm recheck is performed.

[0069] In the embodiment of the present invention, if an erroneous transaction occurs in this transfer, the relevant personnel will be notified to recheck and submit. In addition, if no erroneous transaction occurs in this transfer, the transfer operation to the payee is completed.

[0070] The account early warning method provided by the embodiment of the present invention can determine whether the target account's first transfer information hits the strong rule of erroneous transfer during the transfer process of the target account. If it hits, the second transfer information of the account is predicted to be erroneous, thereby completing the account early warning. Since the amount of information in the second transfer information is higher than the amount of information in the first transfer information, after a small amount of transfer information hits the strong rule, a large amount of transfer information is predicted to be erroneous, which can accurately and quickly identify the erroneous transfer of the account and improve the security of the transfer.

[0071] Based on the account early warning method provided in the above embodiment, the embodiment of the present invention provides a device for executing the above account early warning method. The structural diagram of the device is shown in FIG. Figure 4 As shown, including:

[0072] The strong rule module 101 is used to obtain first transfer information of a target account; if it is determined that the first transfer information hits the configured strong rule, obtain second transfer information of the target account, where the amount of information of the second transfer information is greater than the amount of information of the first transfer information;

[0073] The prediction module 102 is used to predict errors in the second transfer information and perform processing operations on the target account according to the prediction result.

[0074] Optional, strong rule configuration methods include:

[0075] Extracting feature parameters from the first information of the transferor in the transfer dispute event, the feature parameters including amount feature and account feature; analyzing the amount feature and account feature respectively, and determining the transfer amount rule and transfer account rule of the transfer dispute case;

[0076] Accordingly, the strong rule module 101 for determining whether the first transfer information matches a configured strong rule is specifically configured to:

[0077] Determining that the transfer amount in the first transfer information matches the transfer amount rule; or,

[0078] It is determined that the transfer account in the first transfer information matches the transfer account rule.

[0079] Optionally, the prediction module 102 for predicting an error of the second transfer information is specifically configured to:

[0080] The prediction model is retrieved. The prediction model is obtained by training the network parameters of the basic network model with the second information of the transfer parties in the transfer dispute event as training samples, with the goal of making the erroneous prediction results of the training samples close to their erroneous labeling results; the second transfer information is input into the prediction model to obtain the erroneous prediction result of the second transfer information through the prediction model.

[0081] Optionally, the basic network model includes an input layer, a hidden layer and an output layer, and the initial network parameters of the hidden layer are calculated by a genetic algorithm.

[0082] Optionally, the prediction module 102 is further used for:

[0083] In case of an erroneous transfer to the target account, an alarm recheck will be performed.

[0084] It should be noted that the detailed functions of each module in the embodiment of the present invention can be found in the corresponding public part of the above-mentioned account early warning method embodiment, and will not be repeated here.

[0085] Based on the account warning method provided in the above embodiment, the embodiment of the present invention provides an electronic device, see Figure 5 As shown in the structural diagram, the electronic device includes: at least one memory 201 and at least one processor 202; the memory 201 stores a program, and the processor 202 calls the program stored in the memory 201, and the program is used to implement the account early warning method.

[0086] Based on the account early warning method provided in the above embodiment, an embodiment of the present invention provides a storage medium correspondingly, in which computer executable instructions are stored, and the computer executable instructions are used to execute the account early warning method.

[0087] It should be noted that the account early warning method, device, electronic device and storage medium provided by the present invention can be used in the field of artificial intelligence or finance. The above is only an example and does not limit the application field of the account early warning method, device, electronic device and storage medium provided by the present invention.

[0088] The above is a detailed introduction to an account warning method, device, electronic device and storage medium provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

[0089] It should be noted that each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0090] It should also be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device that includes a series of elements is inherent to the elements, or also includes elements inherent to these processes, methods, articles or devices. In the absence of further restrictions, the elements defined by the sentence "including a..." do not exclude the presence of other identical elements in the process, method, article or device that includes the elements.

[0091] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may 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 widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An account early warning method, characterized in that: The method comprises: In response to a transfer instruction of a target account, before transferring the amount of the target account to a payee, obtaining first transfer information of the target account; the first transfer information includes: a transfer amount and a transfer account of the transferor, and the transfer account is an account of the payee; If it is determined that the first transfer information hits the configured strong rule, obtaining second transfer information of the target account, where the amount of information of the second transfer information is greater than the amount of information of the first transfer information; Predicting errors in the second transfer information, and executing a processing operation on the target account according to the prediction result; The step of predicting an error of the second transfer information includes: Retrieving a prediction model, wherein the prediction model is obtained by training network parameters of a basic network model using the second information of the transfer parties in the transfer dispute event as training samples and aiming to make the error prediction results of the training samples close to their error labeling results; Inputting the second transfer information into the prediction model to obtain an error prediction result of the second transfer information through the prediction model; The configuration methods of the strong rule include: Extracting feature parameters from the first information of the transferor in the transfer dispute event, wherein the feature parameters include amount features and account features; Analyze the amount characteristics and the account characteristics respectively, and determine the transfer amount rules and transfer account rules for the transfer dispute event; Accordingly, determining that the first transfer information hits a configured strong rule includes: determining whether the transfer amount in the first transfer information matches the transfer amount rule; or, It is determined that the transfer account in the first transfer information matches the transfer account rule.

2. The method according to claim 1, characterized in that The basic network model includes an input layer, a hidden layer and an output layer, and the initial network parameters of the hidden layer are calculated by a genetic algorithm.

3. The method according to claim 1, characterized in that The performing of the processing operation on the target account according to the prediction result includes: In the event of an erroneous transfer to the target account, an alarm recheck is performed.

4. An account early warning device, characterized in that: The device comprises: A strong rule module, configured to obtain first transfer information of a target account; if it is determined that the first transfer information hits a configured strong rule, obtain second transfer information of the target account, wherein the amount of information of the second transfer information is greater than the amount of information of the first transfer information; A prediction module, used to predict errors in the second transfer information and execute a processing operation on the target account according to the prediction result; The prediction module for predicting an error of the second transfer information is specifically used for: Retrieving a prediction model, wherein the prediction model is obtained by training network parameters of a basic network model using the second information of the transfer parties in the transfer dispute event as training samples, with the goal of making the error prediction results of the training samples close to their error labeling results; inputting the second transfer information into the prediction model, so as to obtain the error prediction results of the second transfer information through the prediction model; The configuration methods of the strong rule include: Extracting characteristic parameters from the first information of the transferor in the transfer dispute event, the characteristic parameters including amount characteristics and account characteristics; analyzing the amount characteristics and the account characteristics respectively to determine the transfer amount rule and the transfer account rule of the transfer dispute event; Accordingly, the strong rule module for determining whether the first transfer information hits a configured strong rule is specifically configured to: determining whether the transfer amount in the first transfer information matches the transfer amount rule; or, It is determined that the transfer account in the first transfer information matches the transfer account rule.

5. An electronic device, characterized in that: The electronic device includes: at least one memory and at least one processor; the memory stores a program, the processor calls the program stored in the memory, and the program is used to implement the account early warning method described in any one of claims 1-3.

6. A storage medium, characterized in that: The storage medium stores computer executable instructions, and the computer executable instructions are used to execute the account early warning method described in any one of claims 1-3.

Images