Financial data sharing safety monitoring method and system based on artificial intelligence

Abstract

The invention relates to the technical field of financial data safety monitoring, in particular to a financial data sharing safety monitoring method and system based on artificial intelligence. The financial data request behavior abnormity intelligent identification model of the user is trained, the requester working range data set of the user initiating the financial data request is evaluated, and whether the financial data request behavior is abnormal or not is judged; and according to the judgment result, judging whether the financial data requested by the user is desensitized and then output to the user, and outputting an alarm signal about financial data request behavior abnormity. The financial data sharing method and device solve the problem that it is difficult to monitor and evaluate whether the data request behaviors of all business personnel are abnormal or not due to the fact that business coupling relations inevitably exist between all clients on the company level, and the safety of financial data sharing is improved to a great extent.

Description

Technical Field

[0001] The present invention relates to the technical field of financial data security monitoring, and in particular to a financial data sharing security monitoring method and system based on artificial intelligence. Background Art

[0002] Before the implementation of financial sharing, the business system, accounting system, capital system, etc. each managed the data security within their own system. After the implementation of financial sharing, the business process, financial process, and capital process are integrated to form a financial sharing process. The relevant data is also integrated through system seamless technology to form a whole data chain of business data, financial data, and capital data, so that the financial sharing management platform forms a data concentration platform for standard data, which can be shared and used by decision-makers, business management, superior authorities, and related parties, promoting the improvement of management effectiveness. The financial sharing management platform has the characteristics of highly concentrated data, huge data volume, and data sharing, and its data security is extremely important. Throughout the life cycle of financial shared data, security technology and security management methods can be used to prevent data leakage, tampering, damage, and loss to ensure the authenticity and integrity of financial shared information; such as data desensitization, which is also an effective means to ensure the security of privacy data in data sharing, distribution, and usage scenarios; However, currently, for data desensitization, different levels of desensitization processing are usually performed according to the permissions of the user requesting the data. However, there is no solution to monitor whether the user's data request behavior is abnormal, resulting in a high security risk of financial data in the desensitization process. For example, at the department level of the same company, personnel belonging to the same business department may not explore and maintain the same customers, but the customers they are responsible for still have personal business secret value to the corresponding business personnel. If there is no solution to monitor whether the user's data request behavior is abnormal, the business secrets of the personal business personnel may still be leaked, resulting in poor security of financial data. However, at the company level, due to the inevitable business coupling relationship between each customer, it is difficult to monitor and evaluate whether the data request behavior of each business personnel is abnormal. How to monitor and evaluate whether the data request behavior of each business personnel is abnormal is a problem that needs to be urgently solved in the existing technology.

[0003] Therefore, the prior art needs to be further developed. Summary of the invention

[0004] The purpose of the present invention is to overcome the above-mentioned technical deficiencies and provide a financial data sharing security monitoring method and system based on artificial intelligence to solve the problems existing in the prior art.

[0005] To achieve the above technical objectives, according to a first aspect of the present invention, the present invention provides a financial data sharing security monitoring method based on artificial intelligence, comprising: S100, constructing a financial data management platform based on the front-end business system, financial system, sharing system and bank-enterprise system, when obtaining a financial data request from a user, obtaining the user's identity information, the user's job title and the keyword field in the financial data requested by the user, and forming a requester's work scope data set according to the user's identity information, the user's job title and the content corresponding to the preset keyword field in the financial data requested by the user; collecting a first preset number of requester's work scope data sets from the user, and dividing the collected requester's work scope data sets into a training data set and a verification data set according to a first preset ratio; S200, using the collected training data set to train the user's financial data request behavior abnormal intelligent recognition model, using the verification data set to optimize the model of the user's financial data request behavior abnormal intelligent recognition model, optimize the model parameters, and complete the training of the user's financial data request behavior abnormal intelligent recognition model; S300. After the model has completed training, it obtains the financial data request from the user again, obtains the user's identity information, the user's job title, and the content corresponding to the preset keyword field in the financial data requested by the user, and forms a target requester's work scope data set. The target requester's work scope data set is input into the user's financial data request behavior abnormal intelligent recognition model, and then determines whether the financial data request behavior is abnormal. According to the judgment result, it is determined whether to desensitize the financial data requested by the user and then output it to the user, and an alarm signal regarding the abnormal financial data request behavior is output.

[0006] Specifically, the obtaining of the identity information of the user, the job title of the user, and the content corresponding to the preset keyword field in the financial data requested by the user includes: Obtain a facial image of the user, identify the facial image of the user, obtain the user's username, and obtain the job title corresponding to the user's username based on the user's username.

[0007] Specifically, the preset keyword field in the financial data requested by the user includes at least one of the following: Customer name, business type, cooperation price, profit, cost, budget.

[0008] Specifically, the method of using the collected training data set to train the user's financial data request behavior abnormal intelligent recognition model, using the verification data set to optimize the model of the user's financial data request behavior abnormal intelligent recognition model, optimizing model parameters, and completing the training of the user's financial data request behavior abnormal intelligent recognition model includes: The training data set is input into the preset network layer in batches for training, and the preset network layer includes a Transformer network layer, and the Transformer network layer is used to predict the working scope data set of the requester of the next financial data request based on the working scope data set of the requester of the current financial data request by forward propagation, and then obtain the predicted loss value; the loss value of the preset network layer is calculated and input into the optimizer for optimization, and the direction in which the parameter gradient of the user's financial data request behavior abnormality intelligent recognition model decreases fastest; the user's financial data request behavior abnormality intelligent recognition model performs back propagation according to the loss value and the parameter gradient of the model, and optimizes the parameters of the user's financial data request behavior abnormality intelligent recognition model.

[0009] Specifically, the method of using the collected training data set to train the user's financial data request behavior abnormal intelligent recognition model, using the verification data set to optimize the model of the user's financial data request behavior abnormal intelligent recognition model, optimizing model parameters, and completing the training of the user's financial data request behavior abnormal intelligent recognition model also includes: After each training, the verification data set is input into the preset network layer of the previous training in batches for model parameter verification, and cyclic training is performed, with the total number of training rounds set to the first preset round; the loss value of the preset network layer is recorded, and it is determined whether the loss value meets the first preset condition, and whether to end the training and output the model parameters based on the judgment result.

[0010] Specifically, the determining whether the loss value satisfies the first preset condition, and determining whether to end the training and output the model parameters according to the determination result, includes: If the loss value meets the first preset condition, the training is terminated and the parameters of the current intelligent recognition model for abnormal financial data request behavior of the user are output; if the loss value does not meet the first preset condition, the training is continued; The first preset condition is that after the second preset round of training, the loss values ​​obtained in the next round of training are greater than or equal to the loss values ​​that occurred during the second preset round of training.

[0011] Specifically, the model output result of the user's financial data request behavior abnormal intelligent identification model includes normal financial data request behavior and abnormal financial data request behavior.

[0012] Specifically, the target requester's work scope data set is input into the user's financial data request behavior abnormal intelligent identification model, and then whether the financial data request behavior is abnormal is determined, and whether the financial data requested by the user is desensitized and then output to the user according to the judgment result, and an alarm signal about the abnormal financial data request behavior is output, including: If the financial data request behavior is determined to be abnormal, the financial data requested by the user is desensitized and then output to the user, and an alarm signal regarding the abnormal financial data request behavior is output.

[0013] Specifically, the target requester's work scope data set is input into the user's financial data request behavior abnormal intelligent identification model, and then whether the financial data request behavior is abnormal is determined, and according to the determination result, whether the financial data requested by the user is desensitized and then output to the user, and an alarm signal about the abnormal financial data request behavior is output, and also includes: If it is determined that the financial data request behavior is normal, the financial data requested by the user is directly output to the user, and a prompt signal indicating that the financial data request behavior is normal is output.

[0014] According to a second aspect of the present invention, there is provided a financial data sharing security monitoring system based on artificial intelligence, comprising: An acquisition module, used to acquire a financial data request from a user; used to acquire the user's identity information, the user's job title, and a keyword field in the financial data requested by the user; The control module is used to build a financial data management platform based on the front-end business system, the financial system, the sharing system and the bank-enterprise system. When a financial data request is obtained from a user, a requester work scope data set is formed according to the identity information of the user, the job title of the user and the content corresponding to the preset keyword field in the financial data requested by the user; a first preset number of requester work scope data sets are collected from the user, and the collected requester work scope data sets are divided into a training data set and a verification data set according to a first preset ratio; the control module is used to use the collected training data set to train the user's financial data request behavior abnormal intelligent recognition model, and use the verification data set to intelligently identify the user's financial data request behavior abnormal The recognition model is optimized, the model parameters are optimized, and the training of the intelligent recognition model for abnormal financial data request behavior of the user is completed; after the model training is completed, the financial data request from the user is obtained again, the identity information of the user, the job title of the user, and the content corresponding to the preset keyword field in the financial data requested by the user are obtained, and a target requester work scope data set is formed, and the target requester work scope data set is input into the intelligent recognition model for abnormal financial data request behavior of the user, so as to determine whether the financial data request behavior is abnormal, and whether the financial data requested by the user is desensitized and output to the user according to the judgment result, and an alarm signal about the abnormal financial data request behavior is output.

[0015] Beneficial effects: The present invention trains an intelligent recognition model for abnormal financial data request behavior of the user, evaluates the work scope data set of the requester of the user who initiates the financial data request, determines whether the financial data request behavior is abnormal, determines whether to desensitize the financial data requested by the user and output it to the user based on the judgment result, and outputs an alarm signal regarding the abnormal financial data request behavior, thereby solving the problem that it is difficult to monitor and evaluate whether the data request behavior of each business person is abnormal due to the inevitable business coupling relationship between various customers at the company level, and greatly improves the security of financial data sharing. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flow chart of a financial data sharing security monitoring method based on artificial intelligence provided in a specific embodiment of the present invention; Figure 2 It is a system composition diagram of the financial data sharing security monitoring system based on artificial intelligence provided in a specific embodiment of the present invention; Figure 3 It is a structural diagram of the intelligent recognition model for abnormal financial data request behavior provided in a specific embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention is clearly and completely described below in conjunction with the accompanying drawings of the present invention. Based on the embodiments of the present invention, other similar embodiments obtained by ordinary technicians in the field without making creative work should all fall within the scope of protection of the present invention. In addition, the directional words mentioned in the following embodiments, such as "up", "down", "left", "right", etc., are only reference to the directions of the accompanying drawings. Therefore, the directional words used are used to illustrate rather than limit the invention.

[0018] The present invention will be further described below in conjunction with the accompanying drawings and preferred embodiments.

[0019] See also Figure 1 The present invention provides a financial data sharing security monitoring method based on artificial intelligence, comprising: S100. Construct a financial data management platform based on the front-end business system, financial system, sharing system and bank-enterprise system. When a financial data request is received from a user, obtain the user's identity information, the user's job title and the keyword field in the financial data requested by the user, and form a requester's work scope data set based on the user's identity information, the user's job title and the content corresponding to the preset keyword field in the financial data requested by the user; collect a first preset number of requester's work scope data sets from the user, and divide the collected requester's work scope data sets into a training data set and a verification data set according to a first preset ratio.

[0020] It should be noted that the steps before step S100 include: A first preset number, a first preset ratio, a first preset round, and a second preset round are preset in the control module.

[0021] It can be understood that the first preset number, the first preset ratio, the first preset round, and the second preset round can be specifically set according to the actual needs of the users of the present invention. The present invention does not limit the specific values ​​of the above parameters, and there is no regularity in setting the specific values ​​of the above parameters, as long as it is applicable to the artificial intelligence-based financial data sharing security monitoring method proposed in the present invention.

[0022] Preferably, the present invention sets the first preset number to 50, the present invention sets the first preset ratio to 8:2, the present invention sets the first preset round to 1000 times, and the present invention sets the second preset round to 100. The above settings can further improve the performance of the model and the efficiency of model training, and greatly improve the accuracy and reliability of the prediction of the intelligent identification model of the abnormal financial data request behavior of the user of the present invention.

[0023] It is understandable that before the user's financial data request behavior abnormal intelligent identification model is officially put into use, the present invention obtains the requester work scope data set of 50 financial data requests made by the same user during normal work, and the relevant management personnel confirm that the 50 financial data request behaviors are normal, that is, the 50 requester work scope data sets are used to train the user's financial data request behavior abnormal intelligent identification model. After the training of the user's financial data request behavior abnormal intelligent identification model is completed, when the user requests financial data again, the user's financial data request behavior abnormal intelligent identification model is used to identify whether the user's request for financial data is abnormal. If it is necessary to monitor the financial data request behavior of multiple users, the financial data request behavior abnormal intelligent identification model of each user can be trained according to the method proposed by the present invention.

[0024] Specifically, the obtaining of the identity information of the user, the job title of the user, and the content corresponding to the preset keyword field in the financial data requested by the user includes: Obtain a facial image of the user, identify the facial image of the user, obtain the user's username, and obtain the job title corresponding to the user's username based on the user's username.

[0025] Specifically, the preset keyword field in the financial data requested by the user includes at least one of the following: Customer name, business type, cooperation price, profit, cost, budget.

[0026] It is understandable that, for example, at the department level of the same company, the personnel belonging to the business department may not have the same customers mined and maintained by each business personnel, but the customers they are responsible for still have personal business secret value for the corresponding business personnel. However, at the company level, although there is an inevitable business coupling relationship between each customer, that is, the same customer may handle different types of business in the company at the same time, the business personnel corresponding to different types of business are still different, that is, the customers, business types, cooperation prices, profits, costs, and budgets that each business personnel is responsible for still have certain fixed rules. The content of the financial data requested by the same business personnel during the work process should also be relatively fixed, and will not suddenly be responsible for other customers or other types of business that originally belonged to others. When the business personnel want to steal the financial data of customers that they are not responsible for, this relatively fixed rule will be broken. Based on this, the present invention designs an intelligent recognition model for abnormal financial data request behavior of the user.

[0027] S200. Use the collected training data set to train the user's financial data request behavior abnormal intelligent recognition model, use the verification data set to optimize the user's financial data request behavior abnormal intelligent recognition model, optimize model parameters, and complete the user's financial data request behavior abnormal intelligent recognition model training.

[0028] Specifically, the method of using the collected training data set to train the user's financial data request behavior abnormal intelligent recognition model, using the verification data set to optimize the model of the user's financial data request behavior abnormal intelligent recognition model, optimizing model parameters, and completing the training of the user's financial data request behavior abnormal intelligent recognition model includes: The training data set is input into the preset network layer in batches for training, and the preset network layer includes a Transformer network layer, and the Transformer network layer is used to predict the working scope data set of the requester of the next financial data request based on the working scope data set of the requester of the current financial data request by forward propagation, and then obtain the predicted loss value; the loss value of the preset network layer is calculated and input into the optimizer for optimization, and the direction in which the parameter gradient of the user's financial data request behavior abnormality intelligent recognition model decreases fastest; the user's financial data request behavior abnormality intelligent recognition model performs back propagation according to the loss value and the parameter gradient of the model, and optimizes the parameters of the user's financial data request behavior abnormality intelligent recognition model.

[0029] It can be understood that Transformer is a deep learning model based on the self-attention mechanism, which consists of a multi-head attention mechanism and a feedforward neural network. It can process sequence data and capture long-distance dependencies. The Transformer model does not contain traditional convolutional layers, but processes each element in the input sequence through an attention layer. The typical structure of Transformer includes an encoder and a decoder, wherein the encoder is composed of multiple identical layers stacked together, and the decoder adds a self-attention layer on the basis of the encoder to process sequence generation tasks. The present invention uses Transformer to supplement global context information, enhances the model's capabilities in feature extraction and global understanding, greatly improves the performance of the model, and greatly reduces the computational cost.

[0030] Specifically, the preset network layer also includes a ResNet network layer, and the ResNet network layer is used to classify and predict the output results based on the collected work scope data set of the requester, and the classification results include normal financial data request behavior and abnormal financial data request behavior.

[0031] It should be noted here that ResNet is a deep convolutional neural network that solves the gradient vanishing problem in deep network training by introducing residual connections. The core component of ResNet is the residual block, which usually contains two or three convolutional layers and a skip connection, which allows the gradient to bypass these convolutional layers directly. The network structure of ResNet can be very deep. These networks extract the user's identity information, the user's job title, and the features of the content corresponding to the preset keyword field in the financial data requested by the user by stacking multiple residual blocks, and classify and predict the output results through global average pooling and fully connected layers. The classification results include normal financial data request behavior and abnormal financial data request behavior.

[0032] See also Figure 3 The present invention combines the ResNet network layer and the Transformer network layer. ResNet is good at extracting the local features of the content corresponding to the user's identity information, the user's job title and the preset keyword field in the financial data requested by the user in the training data set, while Transformer can supplement the global context information. This combination can enhance the model's capabilities in feature extraction and global understanding, greatly improve the performance of the model, and greatly reduce the computational cost.

[0033] It can be understood that the ResNet network layer is used to classify and predict the output results through the requester work scope data set collected in the training data set and the verification data set. The specific steps are: (1) First, the training data set is put into the Backbones network layer of the model. Backbone refers to a series of convolutional layers that constitute the skeleton of the neural network. The main function of these layers is to extract the features of the input data. The Backbone network is usually composed of multiple convolutional layers, pooling layers, and activation functions. It can extract meaningful feature representations from the original data. Backbones is composed of multiple Backbones, and the core network layer in Backbone is the Resnet network layer. After the data set is input into the model, the requester's work scope data set will be input into Backbones for training; (2) When building Backbone, a Resnet network layer was added. After the training data set was input, Resnet began to train layer by layer. The training extracted features from the requester's work scope data set and assisted Transformer in predicting whether the financial data request behavior was abnormal.

[0034] Specifically, the method of using the training data set to train the abnormal intelligent recognition model for the user's financial data request behavior; using the verification data set to optimize the model, optimize the model parameters, and complete the training of the abnormal intelligent recognition model for the user's financial data request behavior, further includes: After each training, the verification data set is input into the preset network layer of the previous training in batches for model parameter verification, and cyclic training is performed, with the total number of training rounds set to the first preset round; the loss value of the preset network layer is recorded, and it is determined whether the loss value meets the first preset condition, and whether to end the training and output the model parameters based on the judgment result.

[0035] Specifically, after each training, the verification data set is input into the preset network layer of the previous training in batches to verify the model parameters, and the cyclic training is performed, including the following methods: When the number of training times is not enough and the model needs to be optimized, the training can only be performed from the beginning. Therefore, the present invention develops a method for continuing the following training based on the previously trained model parameter file; (1) Provide custom parameters, which can be set according to user needs to continue training based on any training parameters; (2) Load the parameters of the custom model file; (3) Put the verification data set into the custom parameter network layer for verification, obtain the loss value and record the model parameters of the model corresponding to the minimum loss value during the training process; (4) Based on this set of model parameters, forward propagation training and back propagation are performed, and the optimizer optimizes the parameters to obtain a new set of data; (5) Put the verification data into the new data network layer for inference to obtain the loss value; (6) Perform cyclic training to obtain the best model and complete model optimization.

[0036] It is understandable that the training method adopted by the present invention is batch training, which means that each time the model parameters are updated, only a part of the samples in the validation data set is used, which is called a batch. The advantages of batch training are that it can reduce memory consumption, speed up training, increase randomness, and facilitate the generalization of the model.

[0037] It should be noted here that, in the present invention, batch_size=2, and the batch_size is the batch size. For example, when batch_size=2, it means that two applicant work scope data sets are selected from the verification data set each time and put into the model for verification.

[0038] Specifically, the determining whether the loss value satisfies the first preset condition, and determining whether to end the training and output the model parameters according to the determination result, includes: If the loss value meets the first preset condition, the training is terminated and the parameters of the current intelligent recognition model for abnormal financial data request behavior of the user are output; if the loss value does not meet the first preset condition, the training is continued; The first preset condition is that after the second preset round of training, the loss values ​​obtained in the next round of training are greater than or equal to the loss values ​​that occurred during the second preset round of training.

[0039] It can be understood that the present invention preferably sets the maximum round of cyclic training to 1000 times, which can effectively ensure the accuracy of the model prediction of the present invention and record the loss value of the model training. The present invention sets the second preset round to 100, that is, when the loss value of the training no longer decreases within 100 rounds of training, the round with the smallest loss value in the 100 rounds of training is saved as the best verification round parameter to generate the best model file. The above setting makes the training rounds no longer rely on manual judgment, but uses the deep learning model for automatic training, which improves the accuracy of the model prediction, and automatically ends the training when the accuracy meets the requirements, which greatly saves the training time, and can also effectively prevent overfitting, which greatly improves the intelligence level of the present invention and the efficiency of model training.

[0040] Furthermore, the specific process of the abnormal intelligent identification model of the user's financial data request behavior of the present invention is as follows: (1) Obtain the latent vector mean mu and logarithmic variance logvar of the user’s financial data request behavior abnormality intelligent identification model in the forward propagation process; (2) Calculate a reasonable loss value representing the training effect: Calculate the divergence: k1=-0.5*(1+logvar-mu^2-(e^logvar)); Calculate the loss between the applicant's work scope data and the predicted value in the validation data set: loss=L1+k1*k1_weight; Among them, loss is the loss value, k1 is the divergence, k1_weight is the proportion of k1, the value is 10, L1 is the absolute difference between the predicted value and the true value, and the calculation formula of L1 is as follows: ; Among them, A is the L1 loss in pytorch, which calculates the average or sum of the absolute differences between two tensors; A has a shape of N*C*H*W, which indicates that it is a four-dimensional tensor, N represents the number of samples (Batch Size), indicating that there are N data points, C represents the number of channels (Channels), H and W represent the height and width respectively, which are usually the two-dimensional spatial dimensions of the input features; P represents the padding value, which is a Boolean type, and the shape of P is c*h*w*1. The dimensions of P are the same as those of A in terms of channel, height, and width, but there is no sample number N, which means that P is a mask or padding information for a feature map of each sample, and the Boolean value of P may be used to mark whether the elements in A are filled (True) or valid (False).

[0041] (3) The loss value when using the validation data set to validate the current model network; (4) If the validation loss of this round is less than that of the previous round, the best model parameters are replaced with the training model of this round; (5) Put the rounds and loss values ​​into the dictionary one by one; (6) Compare the loss value of the current round with the loss values ​​of the 100 rounds before this round (parameters are provided and can be set according to actual needs, the default is 100), that is, traverse the loss values ​​from epoch-100 to epoch-1 (epoch is the training round, and the comparison is performed when the epoch is greater than 100). If the loss values ​​of the current round of epoch are all greater than the loss values ​​from (epoch-100) to (epoch-1), then stop training.

[0042] (7) Save the best model parameters in the cyclic verification as a ckpt model file.

[0043] S300. After the model has completed training, it obtains the financial data request from the user again, obtains the user's identity information, the user's job title, and the content corresponding to the preset keyword field in the financial data requested by the user, and forms a target requester's work scope data set. The target requester's work scope data set is input into the user's financial data request behavior abnormal intelligent recognition model, and then determines whether the financial data request behavior is abnormal. According to the judgment result, it is determined whether to desensitize the financial data requested by the user and then output it to the user, and an alarm signal regarding the abnormal financial data request behavior is output.

[0044] Specifically, the model output result of the user's financial data request behavior abnormal intelligent identification model includes normal financial data request behavior and abnormal financial data request behavior.

[0045] Specifically, the target requester's work scope data set is input into the user's financial data request behavior abnormal intelligent identification model, and then whether the financial data request behavior is abnormal is determined, and whether the financial data requested by the user is desensitized and then output to the user according to the judgment result, and an alarm signal about the abnormal financial data request behavior is output, including: If the financial data request behavior is determined to be abnormal, the financial data requested by the user is desensitized and then output to the user, and an alarm signal regarding the abnormal financial data request behavior is output.

[0046] Specifically, the target requester's work scope data set is input into the user's financial data request behavior abnormal intelligent identification model, and then whether the financial data request behavior is abnormal is determined, and according to the determination result, whether the financial data requested by the user is desensitized and then output to the user, and an alarm signal about the abnormal financial data request behavior is output, and also includes: If it is determined that the financial data request behavior is normal, the financial data requested by the user is directly output to the user, and a prompt signal indicating that the financial data request behavior is normal is output.

[0047] It should be noted here that the present invention trains an intelligent recognition model for abnormal financial data request behavior of the user, evaluates the work scope data set of the requester of the user who initiates the financial data request, determines whether the financial data request behavior is abnormal, determines whether to desensitize the financial data requested by the user and output it to the user based on the judgment result, and outputs an alarm signal regarding the abnormal financial data request behavior, thereby solving the problem that it is difficult to monitor and evaluate whether the data request behavior of each business personnel is abnormal due to the inevitable business coupling relationship between various customers at the company level, and greatly improves the security of financial data sharing.

[0048] See also Figure 2 The present invention provides another embodiment, which provides a financial data sharing security monitoring system based on artificial intelligence, and the financial data sharing security monitoring system based on artificial intelligence includes: The acquisition module 100 is used to acquire a financial data request from a user; and is used to acquire the user's identity information, the user's job title, and a keyword field in the financial data requested by the user; The control module 200 is used to build a financial data management platform based on the front-end business system, the financial system, the sharing system and the bank-enterprise system. When a financial data request is obtained from a user, a requester work scope data set is formed according to the identity information of the user, the job title of the user and the content corresponding to the preset keyword field in the financial data requested by the user; a first preset number of requester work scope data sets are collected from the user, and the collected requester work scope data sets are divided into a training data set and a verification data set according to a first preset ratio; and a financial data request behavior abnormal intelligent recognition model is trained using the collected training data set, and the financial data request behavior abnormal intelligent recognition model is trained using the verification data set. The intelligent recognition model performs model optimization, optimizes model parameters, and completes the training of the intelligent recognition model for abnormal financial data request behavior of the user; after the model is trained, it is used to obtain the financial data request from the user again, obtain the user's identity information, the user's job title, and the content corresponding to the preset keyword field in the financial data requested by the user, form a target requester's work scope data set, input the target requester's work scope data set into the intelligent recognition model for abnormal financial data request behavior of the user, and then determine whether the financial data request behavior is abnormal, determine whether to desensitize the financial data requested by the user and output it to the user based on the judgment result, and output an alarm signal about the abnormal financial data request behavior.

[0049] It can be understood that the present invention trains the intelligent recognition model for abnormal financial data request behavior of the user, evaluates the work scope data set of the requester of the user who initiates the financial data request, determines whether the financial data request behavior is abnormal, determines whether to desensitize the financial data requested by the user and output it to the user based on the judgment result, and outputs an alarm signal regarding the abnormal financial data request behavior, thereby solving the problem that it is difficult to monitor and evaluate whether the data request behavior of each business personnel is abnormal due to the inevitable business coupling relationship between various customers at the company level, and greatly improves the security of financial data sharing.

[0050] In a preferred embodiment, the present invention further provides an electronic device, the electronic device comprising: A memory; and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the method for security monitoring of financial data sharing based on artificial intelligence is implemented. The computer device can be broadly a server, a terminal, or any other electronic device with necessary computing and / or processing capabilities. In one embodiment, the computer device may include a processor, a memory, a network interface, a communication interface, etc. connected via a system bus. The processor of the computer device can be used to provide necessary computing, processing and / or control capabilities. The memory of the computer device may include a non-volatile storage medium and an internal memory. An operating system, a computer program, etc. may be stored in or on the non-volatile storage medium. The internal memory may provide an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface and the communication interface of the computer device can be used to connect and communicate with external devices through a network. When the computer program is executed by the processor, the steps of the method of the present invention are performed.

[0051] The present invention may be implemented as a computer-readable storage medium having a computer program stored thereon, which causes the steps of the method of an embodiment of the present invention to be executed when executed by a processor. In one embodiment, the computer program is distributed on a plurality of computer devices or processors coupled to a network so that the computer program is stored, accessed and executed in a distributed manner by one or more computer devices or processors. A single method step / operation, or two or more method steps / operations, may be performed by a single computer device or processor or by two or more computer devices or processors. One or more method steps / operations may be performed by one or more computer devices or processors, and one or more other method steps / operations may be performed by one or more other computer devices or processors. One or more computer devices or processors may perform a single method step / operation, or perform two or more method steps / operations.

[0052] It will be appreciated by a person skilled in the art that the method steps of the present invention may be performed by instructing related hardware such as a computer device or a processor through a computer program, and the computer program may be stored in a non-temporary computer-readable storage medium, which causes the steps of the present invention to be performed when the computer program is executed. Depending on the circumstances, any reference to memory, storage, database, or other media herein may include non-volatile and / or volatile memory. Examples of non-volatile memory include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid state disk, etc. Examples of volatile memory include random access memory (RAM), external cache memory, etc.

[0053] It can be understood that the present invention trains the intelligent recognition model for abnormal financial data request behavior of the user, evaluates the work scope data set of the requester of the user who initiates the financial data request, determines whether the financial data request behavior is abnormal, determines whether to desensitize the financial data requested by the user and output it to the user based on the judgment result, and outputs an alarm signal regarding the abnormal financial data request behavior, thereby solving the problem that it is difficult to monitor and evaluate whether the data request behavior of each business personnel is abnormal due to the inevitable business coupling relationship between various customers at the company level, and greatly improves the security of financial data sharing.

[0054] The various technical features described above can be combined arbitrarily. Although all possible combinations of these technical features are not described, any combination of these technical features should be considered to be covered by this specification as long as there is no contradiction in such combination.

[0055] The specific implementation of the present invention described above does not constitute a limitation on the protection scope of the present invention. Any other corresponding changes and modifications made based on the technical concept of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A financial data sharing security monitoring method based on artificial intelligence, characterized in that: The method comprises: S100, constructing a financial data management platform based on the front-end business system, financial system, sharing system and bank-enterprise system, when obtaining a financial data request from a user, obtaining the user's identity information, the user's job title and the keyword field in the financial data requested by the user, and forming a requester's work scope data set according to the user's identity information, the user's job title and the content corresponding to the preset keyword field in the financial data requested by the user; collecting a first preset number of requester's work scope data sets from the user, and dividing the collected requester's work scope data sets into a training data set and a verification data set according to a first preset ratio; S200, using the collected training data set to train the user's financial data request behavior abnormal intelligent recognition model, using the verification data set to optimize the model of the user's financial data request behavior abnormal intelligent recognition model, optimize the model parameters, and complete the training of the user's financial data request behavior abnormal intelligent recognition model; S300. After the model has completed training, it obtains the financial data request from the user again, obtains the user's identity information, the user's job title, and the content corresponding to the preset keyword field in the financial data requested by the user, and forms a target requester's work scope data set. The target requester's work scope data set is input into the user's financial data request behavior abnormal intelligent recognition model, and then determines whether the financial data request behavior is abnormal. According to the judgment result, it is determined whether to desensitize the financial data requested by the user and then output it to the user, and an alarm signal regarding the abnormal financial data request behavior is output.

2. The method for financial data sharing security monitoring based on artificial intelligence according to claim 1 is characterized in that: The obtaining of the identity information of the user, the job title of the user, and the content corresponding to the preset keyword field in the financial data requested by the user includes: Obtain a facial image of the user, identify the facial image of the user, obtain the user's username, and obtain the job title corresponding to the user's username based on the user's username.

3. The method for financial data sharing security monitoring based on artificial intelligence according to claim 1 is characterized in that: The preset keyword field in the financial data requested by the user includes at least one of the following: Customer name, business type, cooperation price, profit, cost, budget.

4. The method for financial data sharing security monitoring based on artificial intelligence according to claim 1 is characterized in that: The method of using the collected training data set to train the abnormal intelligent recognition model for the user's financial data request behavior, using the verification data set to optimize the model, optimizing the model parameters, and completing the training of the abnormal intelligent recognition model for the user's financial data request behavior includes: The training data set is input into the preset network layer in batches for training, and the preset network layer includes a Transformer network layer, and the Transformer network layer is used to predict the working scope data set of the requester of the next financial data request based on the working scope data set of the requester of the current financial data request by forward propagation, and then obtain the predicted loss value; the loss value of the preset network layer is calculated and input into the optimizer for optimization, and the direction in which the parameter gradient of the user's financial data request behavior abnormality intelligent recognition model decreases fastest; the user's financial data request behavior abnormality intelligent recognition model performs back propagation according to the loss value and the parameter gradient of the model, and optimizes the parameters of the user's financial data request behavior abnormality intelligent recognition model.

5. The method for financial data sharing security monitoring based on artificial intelligence according to claim 4 is characterized in that: The method of using the collected training data set to train the abnormal intelligent recognition model for the user's financial data request behavior, using the verification data set to optimize the model, optimizing the model parameters, and completing the training of the abnormal intelligent recognition model for the user's financial data request behavior, further includes: After each training, the verification data set is input into the preset network layer of the previous training in batches for model parameter verification, and cyclic training is performed, with the total number of training rounds set to the first preset round; the loss value of the preset network layer is recorded, and it is determined whether the loss value meets the first preset condition, and whether to end the training and output the model parameters based on the judgment result.

6. The method for financial data sharing security monitoring based on artificial intelligence according to claim 5 is characterized in that: The determining whether the loss value satisfies the first preset condition, and determining whether to end the training and output the model parameters according to the determination result, includes: If the loss value meets the first preset condition, the training is terminated and the parameters of the current intelligent recognition model for abnormal financial data request behavior of the user are output; if the loss value does not meet the first preset condition, the training is continued; The first preset condition is that after the second preset round of training, the loss values ​​obtained in the next round of training are greater than or equal to the loss values ​​that occurred during the second preset round of training.

7. The method for financial data sharing security monitoring based on artificial intelligence according to claim 1 is characterized in that: The model output result of the user's financial data request behavior abnormal intelligent identification model includes normal financial data request behavior and abnormal financial data request behavior.

8. The method for financial data sharing security monitoring based on artificial intelligence according to claim 7 is characterized in that: The target requester's work scope data set is input into the user's financial data request behavior abnormal intelligent identification model, and then whether the financial data request behavior is abnormal is determined, and whether the financial data requested by the user is desensitized and then output to the user according to the judgment result, and an alarm signal about the abnormal financial data request behavior is output, including: If the financial data request behavior is determined to be abnormal, the financial data requested by the user is desensitized and then output to the user, and an alarm signal regarding the abnormal financial data request behavior is output.

9. The method for financial data sharing security monitoring based on artificial intelligence according to claim 7 is characterized in that: The target requester's work scope data set is input into the user's financial data request behavior abnormal intelligent identification model, and then whether the financial data request behavior is abnormal is determined, and whether the financial data requested by the user is desensitized and then output to the user according to the judgment result, and an alarm signal about the abnormal financial data request behavior is output, and also includes: If it is determined that the financial data request behavior is normal, the financial data requested by the user is directly output to the user, and a prompt signal indicating that the financial data request behavior is normal is output.

10. A financial data sharing security monitoring system based on artificial intelligence, characterized in that: include: The acquisition module is used to obtain financial data requests from users; Used to obtain the user's identity information, the user's job title, and the keyword fields in the financial data requested by the user; A control module is used to build a financial data management platform based on a front-end business system, a financial system, a sharing system and a bank-enterprise system. When a financial data request is obtained from a user, a requester work scope data set is formed according to the identity information of the user, the job title of the user and the content corresponding to the preset keyword field in the financial data requested by the user; a first preset number of requester work scope data sets are collected from the user, and the collected requester work scope data sets are divided into a training data set and a verification data set according to a first preset ratio; It is used to train the user's financial data request behavior abnormal intelligent recognition model using the collected training data set, and to optimize the model of the user's financial data request behavior abnormal intelligent recognition model using the verification data set, optimize the model parameters, and complete the training of the user's financial data request behavior abnormal intelligent recognition model; it is used to obtain the financial data request from the user again after the model training is completed, obtain the user's identity information, the user's job title, and the content corresponding to the preset keyword field in the financial data requested by the user, form a target requester's work scope data set, input the target requester's work scope data set into the user's financial data request behavior abnormal intelligent recognition model, and then determine whether the financial data request behavior is abnormal, determine whether to desensitize the financial data requested by the user and output it to the user based on the judgment result, and output an alarm signal about the abnormal financial data request behavior.

Images