Data risk assessment method and device, equipment, storage medium and program product

Abstract

The invention provides a data risk assessment method, which can be used in the financial science and technology field or other fields, and comprises the following steps: obtaining to-be-assessed credit data; based on a risk assessment model deployed on the target system, the credit data to be assessed are assessed to obtain an assessment result, and the risk assessment model is obtained through pre-training; and explaining the evaluation result by adopting an interpretability technology to obtain an interpretation result, and carrying out visualization processing on the interpretation result. The invention further provides a data risk assessment device and equipment, a storage medium and a program product.

Description

Technical Field

[0001] The present disclosure relates to the field of financial technology, and more specifically to a data risk assessment method, device, equipment, storage medium and program product. Background Art

[0002] At present, the methods of bank credit loan risk assessment are mainly divided into: one is the traditional manual verification method, which judges the customer's credit status and repayment ability based on the borrower's credit report, income verification, mortgage value assessment, and understanding of the customer's work and family situation. The other is the use of new computer technologies, including machine learning, for assessment and data analysis. Machine learning models can make predictions based on historical data and models to help banks identify potential risk points.

[0003] For the above two existing technologies, the traditional manual verification method requires manual participation in obtaining and verifying data, which has certain limitations in terms of efficiency and real-time performance. For the method of using machine learning for risk assessment, the generalization ability of the machine learning model is limited, and it cannot fully balance a large amount of bank customer data, which is prone to overfitting; in addition, as the complexity of the combination of machine learning models increases, the training and maintenance costs also increase greatly. The cumbersome feature engineering of the machine learning model will also reduce the efficiency of the model, and its interpretability will also be reduced accordingly. It is not suitable for application in the actual bank customer loan risk prediction, which limits the decision-making of credit personnel. Summary of the invention

[0004] In view of the above problems, the present disclosure provides a data risk assessment method, apparatus, device, storage medium and program product.

[0005] According to a first aspect of the present disclosure, a data risk assessment method is provided, comprising: obtaining credit data to be assessed; assessing the credit data to be assessed based on a risk assessment model deployed on a target system to obtain an assessment result, wherein the risk assessment model is obtained through pre-training; interpreting the assessment result using interpretability technology to obtain an interpretation result, and visualizing the interpretation result.

[0006] According to an embodiment of the present disclosure, a risk assessment model is obtained through pre-training, including: obtaining original data; dividing and preprocessing the original data, wherein the original data is divided into a data training set, a data test set, and a data verification set; constructing a risk assessment model, and using the preprocessed data training set to train the risk assessment model to obtain a trained risk assessment model.

[0007] According to an embodiment of the present disclosure, the risk assessment model is obtained through pre-training, and also includes: using the preprocessed data test set to test the trained risk assessment model to obtain a predictive ability index of the risk assessment model; based on the predictive ability index, using the data verification set to adjust the parameters of the risk assessment model to optimize the risk assessment model.

[0008] According to an embodiment of the present disclosure, the risk assessment model is deployed on a target system, and the deployment method includes: converting the trained risk assessment model into a target format according to the requirements of the target system; packaging the risk assessment model converted into the target format into an image, and deploying it on the target system.

[0009] According to an embodiment of the present disclosure, the risk assessment model is deployed on the target system, and the deployment method also includes: performing unit testing, integration testing and load testing on the target system to verify the performance of the target system and the accuracy of the risk assessment model in the target system; monitoring the performance of the target system and the accuracy of the risk assessment model in the target system.

[0010] According to an embodiment of the present disclosure, an interpretability technology is used to interpret the evaluation results to obtain an explanation result, including: obtaining the evaluation results and defining the evaluation results as target samples; generating multiple neighborhood samples near the feature space of the target samples; inputting the neighborhood samples into the original black box model to obtain the prediction results of each sample; based on the features of the neighborhood samples and the prediction results, using linear regression to fit a local explanation model to determine the influence of each feature on the model prediction.

[0011] According to an embodiment of the present disclosure, the interpretation results are visualized, including: presenting the interpretation results in a graphical manner to show the impact of different features on the evaluation results; analyzing the interaction between the features based on the interpretation results to identify the main influencing factors.

[0012] According to an embodiment of the present disclosure, the raw data is preprocessed, including: cleaning the raw data; deduplicating, filling missing values, converting data types, and desensitizing the cleaned raw data; and extracting features from the desensitized raw data based on business needs.

[0013] The second aspect of the present disclosure provides a data risk assessment device, including: a data acquisition module, used to acquire credit data to be assessed; a data risk assessment module, used to assess the credit data to be assessed based on a risk assessment model deployed on a target system, and obtain an assessment result, wherein the risk assessment model is obtained through pre-training; an interpretation module, used to interpret the assessment result using interpretability technology, obtain an interpretation result, and visualize the interpretation result.

[0014] According to an embodiment of the present disclosure, the data risk assessment module includes a model training submodule and a model optimization submodule, wherein the model training submodule includes: a raw data acquisition unit, used to acquire raw data; a preprocessing unit, used to divide and preprocess the raw data, wherein the raw data is divided into a data training set, a data test set and a data verification set; a model training unit, used to construct a risk assessment model, and use the preprocessed data training set to train the risk assessment model to obtain a trained risk assessment model.

[0015] According to an embodiment of the present disclosure, the model optimization submodule includes: a model testing unit, which is used to test the trained risk assessment model using a preprocessed data test set to obtain a predictive ability index of the risk assessment model; and a model adjustment unit, which is used to adjust the parameters of the risk assessment model based on the predictive ability index and use the data verification set to optimize the risk assessment model.

[0016] A third aspect of the present disclosure provides an electronic device, comprising: one or more processors; and a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the above method.

[0017] The fourth aspect of the present disclosure further provides a computer-readable storage medium having a computer program or instructions stored thereon, which implements the steps of the above method when the above computer program or instructions are executed by a processor.

[0018] The fifth aspect of the present disclosure further provides a computer program product, including a computer program or instructions, which implement the steps of the above method when the above computer program or instructions are executed by a processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The above contents and other objects, features and advantages of the present disclosure will become more apparent through the following description of the embodiments of the present disclosure with reference to the accompanying drawings, in which:

[0020] Figure 1 The application scenario diagram of the data risk assessment method according to the embodiment of the present disclosure is schematically shown;

[0021] Figure 2 A flowchart of a data risk assessment method according to an embodiment of the present disclosure is schematically shown;

[0022] Figure 3 A flowchart schematically shows a risk assessment model obtained through pre-training according to an embodiment of the present disclosure;

[0023] Figure 4 A flowchart for optimizing a risk assessment model according to an embodiment of the present disclosure is schematically shown;

[0024] Figure 5 A flowchart for explaining the evaluation results according to an embodiment of the present disclosure is schematically shown;

[0025] Figure 6 A schematic diagram of a data risk assessment device according to an embodiment of the present disclosure is shown; and

[0026] Figure 7 A block diagram of an electronic device suitable for implementing a data risk assessment method according to an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0027] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present disclosure. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present disclosure. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.

[0028] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "comprise", "include", etc. used herein indicate the existence of features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0029] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0030] When using expressions such as "at least one of A, B, and C, etc.", they should generally be interpreted according to the meaning of the expression commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0031] It should be noted that the data risk assessment method and device disclosed herein can be used in the field of financial technology, and can also be used in any field other than the field of financial technology. The application field of the data risk assessment method and device disclosed herein is not limited.

[0032] In the technical solution of the present disclosure, the user information (including but not limited to user personal information, user image information, user device information, such as location information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved are all information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with relevant laws, regulations and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0033] In the scenario of using personal information for automated decision-making, the methods, devices, and systems provided by the embodiments of the present disclosure provide users with corresponding operation portals for users to choose to agree or reject the automated decision-making results; if the user chooses to reject, the expert decision-making process will be entered. The expression "automated decision-making" here refers to the activity of automatically analyzing and evaluating a person's behavioral habits, interests and hobbies, or economic, health, credit status, etc. through computer programs, and making decisions. The expression "expert decision-making" here refers to the activity of making decisions by people who specialize in a certain field, have specialized experience, knowledge and skills, and have reached a certain level of professionalism.

[0034] An embodiment of the present disclosure provides a data risk assessment method, which includes: obtaining credit data to be assessed; assessing the credit data to be assessed based on a risk assessment model deployed on a target system to obtain an assessment result, wherein the risk assessment model is obtained through pre-training; interpreting the assessment result using interpretability technology to obtain an interpretation result, and visualizing the interpretation result.

[0035] Through the embodiments of the present disclosure, through pre-trained risk assessment models and systematic data processing, the ability to accurately analyze credit data to be assessed can be effectively improved, thereby providing a more reliable basis for decision-making; in addition, the use of interpretability technology to conduct in-depth analysis of the assessment results makes the model's judgment process transparent, and credit personnel can understand why the model gives a specific assessment result. Through the visualization of the interpretation results, the impact of each feature on the assessment result and their interaction can be clearly displayed, helping credit personnel identify the main influencing factors and providing guidance for subsequent data processing and decision-making.

[0036] Through the embodiments of the present disclosure, the method can not only provide efficient and accurate results in data risk assessment, but also help credit personnel to have a deep understanding, ultimately achieving the purpose of reducing risks and improving decision-making quality.

[0037] Figure 1 The application scenario diagram of the data risk assessment method according to an embodiment of the present disclosure is schematically shown.

[0038] like Figure 1 As shown, the application scenario 100 according to this embodiment may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 is used to provide a medium for communication links between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or optical fiber cables, etc.

[0039] The user can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only for example).

[0040] The first terminal device 101, the second terminal device 102, and the third terminal device 103 may be various electronic devices having display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0041] The server 105 may be a server that provides various services, such as a background management server (only as an example) that provides support for websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103. The background management server may analyze and process the received data such as user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal device.

[0042] It should be noted that the data risk assessment method provided in the embodiment of the present disclosure can generally be executed by the server 105. Accordingly, the data risk assessment device provided in the embodiment of the present disclosure can generally be set in the server 105. The data risk assessment method provided in the embodiment of the present disclosure can also be executed by a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105. Accordingly, the data risk assessment device provided in the embodiment of the present disclosure can also be set in a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105.

[0043] It should be understood that Figure 1The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to the implementation requirements.

[0044] The following will be based on Figure 1 The scene described by Figure 2~Figure 5 The data risk assessment method of the disclosed embodiment is described in detail.

[0045] Figure 2 The flowchart of the data risk assessment method according to the embodiment of the present disclosure is schematically shown.

[0046] like Figure 2 As shown, in an embodiment of the present disclosure, the data risk assessment method specifically includes operations S210 to S230.

[0047] In operation S210, credit data to be evaluated is obtained.

[0048] For example, in the scenario of bank credit loans, in order to complete the loan business, it is necessary to conduct a risk assessment on the customer's relevant data. First, obtain the customer's authorization to obtain the credit data to be evaluated, and then obtain the customer's credit data to be evaluated. For example, the credit data to be evaluated includes the borrower's personal information, credit history, credit score and other information.

[0049] In operation S220, the credit data to be evaluated is evaluated based on the risk evaluation model deployed on the target system to obtain an evaluation result, wherein the risk evaluation model is obtained through pre-training.

[0050] In the embodiment of the present disclosure, in the scenario of bank credit loans, the target system may be, for example, a bank credit system. After acquiring the credit data to be evaluated, the bank credit system inputs the credit data to be evaluated into a pre-trained risk assessment model to perform risk assessment and obtain an assessment result.

[0051] It should be noted that the risk assessment model is obtained through training and optimization. The training and optimization of the risk assessment model are explained below.

[0052] Figure 3 The flowchart schematically shows a risk assessment model obtained through pre-training according to an embodiment of the present disclosure.

[0053] like Figure 3 As shown, in an embodiment of the present disclosure, the risk assessment model is obtained through pre-training, which specifically includes operations S310 to S330.

[0054] In operation S310, original data is acquired.

[0055] In operation S320, the original data is divided and preprocessed, wherein the original data is divided into a data training set, a data test set, and a data verification set.

[0056] In the embodiments of the present disclosure, before constructing and training the risk assessment model, the original data is first obtained, and the original data may also be various customer-related credit data in the historical database that has obtained customer authorization. The credit data may include, for example, the borrower's personal information, credit history, credit score and other information.

[0057] In the embodiments of the present disclosure, the customer-related credit data is further preprocessed, including: cleaning the original data; deduplicating the cleaned original data, filling missing values ​​and converting data types, and desensitizing the data; based on business needs, extracting features from the desensitized original data.

[0058] Specifically, these customer-related credit data are cleaned, deduplicated, missing values ​​are filled, data types are converted, etc. to ensure data quality and consistency. Then, the data is desensitized while protecting privacy. Data desensitization is a process of replacing sensitive data with meaningless or synthetic data to protect the privacy and security of the original data. For example, deep learning models such as recurrent neural networks (RNNs) are used to learn semantic and contextual information in text, accurately identify sensitive information in customer-related credit data, and delete sensitive information for desensitization. After desensitization, the data is verified to ensure the accuracy of the desensitized information operation.

[0059] After the above steps, a clean and structured target data set is obtained, which will be used for subsequent data analysis and modeling. In the embodiment of the present disclosure, the target data set is also divided into a data training set, a data test set and a data verification set.

[0060] Furthermore, after the data set is divided, feature extraction is performed on the data training set, data test set and data validation set respectively.

[0061] Exemplarily, a target feature is determined in advance, such as a credit rating time feature in customer-related credit data, a historical credit maximum feature, etc. Then, a recurrent neural network is used to extract features from the data training set, the data test set, and the data validation set, respectively, so that the extracted features of the data training set, the data test set, and the data validation set remain consistent.

[0062] Through the embodiments of the present disclosure, feature extraction is performed through a recurrent neural network, and a high-level nonlinear transformation is performed, so that the subsequent risk assessment model can automatically learn features suitable for specific tasks. This process reduces the complexity of manual features and improves the performance of the model in various applications.

[0063] In operation S330, a risk assessment model is constructed, and the risk assessment model is trained using the preprocessed data training set to obtain a trained risk assessment model.

[0064] Specifically, first select a suitable deep neural network model, such as a recurrent neural network (RNN) to build a risk assessment model, and then design the input and output of the model, taking customer-related credit data as input and the assessment results as output.

[0065] Furthermore, the risk assessment model is trained based on the data training set that has completed the target feature extraction. During the training process, the network parameters are adjusted and regularization is used to prevent overfitting and improve generalization ability. At the same time, the model is verified using the data validation set to evaluate the performance of the model.

[0066] Through pre-trained risk assessment models and systematic data processing, the ability to accurately analyze the credit data to be assessed can be effectively improved, thus providing a more reliable basis for decision-making.

[0067] Figure 4 A flowchart for optimizing a risk assessment model according to an embodiment of the present disclosure is schematically shown.

[0068] like Figure 4 As shown, in an embodiment of the present disclosure, the risk assessment model is optimized, specifically including operations S410 to S420.

[0069] In operation S410, the trained risk assessment model is tested using the preprocessed data test set to obtain a prediction capability index of the risk assessment model.

[0070] Exemplarily, the preprocessed data test set is input into the trained risk assessment model for testing, and the model's prediction ability indicators such as accuracy, recall rate, and F1 value are calculated.

[0071] In operation S420, based on the prediction capability indicator, the parameters of the risk assessment model are adjusted using the data validation set to optimize the risk assessment model.

[0072] For example, the trained model is then applied to the preprocessed validation set to obtain the prediction results of the validation set, and the prediction ability index is recalculated based on the prediction results of the validation set. The performance of the model is analyzed based on the evaluation results of the validation set, and the model parameters are adjusted. For example, the model is optimized by improving and adjusting the network structure, changing the activation function, etc.

[0073] For example, if the recall rate on the validation set is low, you may need to adjust the model threshold or add certain features; and use grid search or random search methods to tune the model hyperparameters, such as the depth of the tree, the minimum number of sample splits, the learning rate, etc.

[0074] Through the embodiments of the present disclosure, the trained risk assessment model is tested and optimized. By using the test set to evaluate the model's predictive ability indicators and adjusting the model parameters based on the feedback of the validation set, the accuracy and reliability of risk assessment can be significantly improved.

[0075] Based on the above risk assessment model, by training on large-scale data sets, deep neural networks can learn general rules and patterns, thereby accurately assessing new data, which makes it more adaptable and scalable in credit risk assessment.

[0076] In the embodiment of the present disclosure, in the scenario of bank credit loans, the credit data to be evaluated is evaluated based on the risk assessment model deployed on the bank credit system to obtain the evaluation result. That is, the optimized model is deployed to the bank credit system for real-time prediction of credit risk. At the same time, the performance of the model is monitored, and the model is regularly updated and maintained to ensure its prediction accuracy and stability.

[0077] The following is a detailed description of the process of deploying the risk assessment model on the target system.

[0078] In an embodiment of the present disclosure, the risk assessment model is deployed on a target system, and the deployment method includes: converting the trained risk assessment model into a target format according to the requirements of the target system; packaging the risk assessment model converted into the target format into an image, and deploying it on the target system.

[0079] Furthermore, after the risk assessment model is deployed on the target system, i.e., the bank credit system, the target system is subjected to unit testing, integration testing, and load testing to verify the performance of the target system and the accuracy of the risk assessment model in the target system; and to monitor the performance of the target system and the accuracy of the risk assessment model in the target system.

[0080] Through the embodiments of the present disclosure, flexible model deployment can be performed, and the compatibility and effectiveness of the model in different environments can be ensured by format conversion and packaging that adapts to the requirements of the target system.

[0081] In operation S230, the evaluation result is explained using an explainability technology to obtain an explanation result, and the explanation result is visualized.

[0082] In the embodiment of the present disclosure, after the risk assessment model deployed in the bank credit system completes the assessment of the credit data to be assessed, the assessment result is obtained, and the interpretability technology is used to interpret the assessment result to obtain the interpretation result, and the interpretation result is visualized.

[0083] Figure 5 The flowchart for explaining the evaluation result according to the embodiment of the present disclosure is schematically shown.

[0084] like Figure 5 As shown, in an embodiment of the present disclosure, an interpretability technology is used to interpret the evaluation result to obtain an interpretation result, which specifically includes operations S510 to S540.

[0085] In operation S510, an evaluation result is acquired and defined as a target sample.

[0086] In operation S520 , a plurality of neighborhood samples are generated near a feature space of the target sample.

[0087] In operation S530, the neighborhood samples are input into the original black box model to obtain a prediction result for each sample.

[0088] In operation S540, based on the features of the neighborhood samples and the prediction results, a local explanation model is fitted using linear regression to determine the influence of each feature on the model prediction.

[0089] In the disclosed embodiment, the LIME interpretability technology is used to explain the evaluation results of the risk assessment model to understand the features that have the greatest impact on the evaluation results.

[0090] Specifically, after obtaining the evaluation result, the evaluation result is used as the target sample to be explained. Further, the original sample is perturbed using the selected features to generate multiple neighborhood samples. It should be noted that the perturbation can be random, usually by adding or subtracting a small amount of features, and these perturbed samples should cover the feature space of the original sample as much as possible.

[0091] Furthermore, the generated perturbation samples are predicted using the original black box model to obtain the prediction results of each perturbation sample, and the characteristic values ​​of these perturbation samples and the corresponding prediction outputs are recorded.

[0092] Furthermore, based on the prediction results of the perturbed samples and the feature values ​​of the original samples, a simple interpretable model is constructed to fit these perturbed samples. This simple model is used to approximate the behavior of the original black box model near the sample. Then, the parameters of the local linear model (such as regression coefficients) are analyzed to determine the contribution of each feature to the model prediction.

[0093] Furthermore, in an embodiment of the present disclosure, the interpretation results are visualized, including: presenting the interpretation results in a graphical manner to show the impact of different features on the evaluation results; analyzing the interaction between the features based on the interpretation results to identify the main influencing factors.

[0094] Specifically, the interpretation results of LIME are presented in a graphical way, such as a feature importance bar chart, to show the impact of different features on the final prediction results. The interactions between features are analyzed based on the results of LIME to identify the main influencing factors.

[0095] Through the embodiments of the present disclosure, interpretability technology is used to interpret the prediction results of the model and understand the features that have the greatest impact on the prediction results. The interpretation results are presented to bank loan personnel in a visual manner to improve the decision-making ability of bank loan personnel. Interpretability technology provides an in-depth understanding of the model decision-making process, allowing relevant staff to more clearly understand the key factors that affect credit decisions, and increasing the transparency and trustworthiness of decision-making.

[0096] Based on the above data risk assessment method, in view of the strong dependence of traditional technology on manual labor, through the embodiments of the present invention, a deep learning model with higher accuracy is trained using large amounts of data, which can predict the degree of bank lending risk based on key credit indicators of customers. The optimized model can be deployed in the bank's credit system, thereby improving verification autonomy and reducing manpower consumption.

[0097] Based on the above data risk assessment method, in view of the limitations of the generalization ability of machine learning models and the complexity of feature engineering, through the embodiments of the present invention, a deep neural network with strong generalization ability that can automatically extract features is used to identify and desensitize customer sensitive information, and automatic learning and extraction of complex data features are performed in the computing system to achieve higher and more effective prediction accuracy. It also combines explainability technology to explain the reasons and basis for the prediction results of the model, which is helpful for credit personnel to actually use and understand.

[0098] Based on the above data risk assessment method, the present disclosure also provides a data risk assessment device. Figure 6 The device is described in detail.

[0099] Figure 6 The structure block diagram of the data risk assessment device according to an embodiment of the present disclosure is schematically shown.

[0100] like Figure 6 As shown, the data risk assessment device 600 of this embodiment includes a data acquisition module 610 , a data risk assessment module 620 and an interpretation module 630 .

[0101] The data acquisition module 610 is used to acquire the credit data to be evaluated. In one embodiment, the data acquisition module 610 can be used to perform the operation S210 described above, which will not be described in detail here.

[0102] The data risk assessment module 620 is used to assess the credit data to be assessed based on the risk assessment model deployed on the target system to obtain an assessment result, wherein the risk assessment model is obtained through pre-training. In one embodiment, the data risk assessment module 620 can be used to perform the operation S220 described above, which will not be repeated here.

[0103] The explanation module 630 is used to explain the evaluation result using an explainability technology, obtain an explanation result, and visualize the explanation result. In one embodiment, the explanation module 630 can be used to perform the operation S230 described above, which will not be described in detail here.

[0104] In an embodiment of the present disclosure, the data risk assessment module 620 includes a model training submodule and a model optimization submodule, wherein the model training submodule includes a raw data acquisition unit, a preprocessing unit and a model training unit.

[0105] The original data acquisition unit is used to acquire the original data. In one embodiment, the original data acquisition unit can be used to perform the operation S310 described above, which will not be described in detail here.

[0106] The preprocessing unit is used to divide and preprocess the original data, wherein the original data is divided into a data training set, a data test set and a data verification set. In one embodiment, the preprocessing unit can be used to perform the operation S320 described above, which will not be repeated here.

[0107] The model training unit is used to construct a risk assessment model and train the risk assessment model using the preprocessed data training set to obtain a trained risk assessment model. In one embodiment, the model training unit can be used to perform the operation S330 described above, which will not be described in detail here.

[0108] In an embodiment of the present disclosure, the model optimization submodule includes a model testing unit and a model adjustment unit.

[0109] The model testing unit is used to test the trained risk assessment model using the preprocessed data test set to obtain the prediction capability index of the risk assessment model. In one embodiment, the model testing unit can be used to perform the operation S410 described above, which will not be repeated here.

[0110] The model adjustment unit is used to adjust the parameters of the risk assessment model based on the prediction capability index and the data validation set to optimize the risk assessment model. In one embodiment, the model adjustment unit can be used to perform the operation S420 described above, which will not be described in detail here.

[0111] In an embodiment of the present disclosure, the explanation module 630 includes an explanation first submodule and an explanation second submodule, wherein the explanation first submodule includes an evaluation result definition unit, a neighborhood sample generation unit, a prediction result unit and a local explanation model generation unit.

[0112] The evaluation result definition unit is used to obtain the evaluation result and define the evaluation result as the target sample. In one embodiment, the evaluation result definition unit can be used to perform the operation S510 described above, which will not be described in detail here.

[0113] The neighborhood sample generating unit is used to generate a plurality of neighborhood samples near the feature space of the target sample. In one embodiment, the neighborhood sample generating unit can be used to perform the operation S520 described above, which will not be described in detail here.

[0114] The prediction result unit is used to input the neighborhood samples into the original black box model to obtain the prediction result of each sample. In one embodiment, the prediction result unit can be used to perform the operation S530 described above, which will not be described in detail here.

[0115] The local explanation model generation unit is used to fit a local explanation model using linear regression based on the features and prediction results of the neighborhood samples to determine the influence of each feature on the model prediction. In one embodiment, the local explanation model generation unit can be used to perform the operation S540 described above, which will not be repeated here.

[0116] In an embodiment of the present disclosure, the second submodule is explained to include a graphic display unit and a recognition unit.

[0117] The graphic display unit is used to present the interpretation results in a graphical manner and show the impact of different features on the evaluation results.

[0118] The identification unit is used to analyze the interactions between features and identify the main influencing factors based on the interpretation results.

[0119] According to an embodiment of the present disclosure, any multiple modules of the data acquisition module 610, the data risk assessment module 620 and the interpretation module 630 can be combined into one module for implementation, or any one of the modules can be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules can be combined with at least part of the functions of other modules and implemented in one module. According to an embodiment of the present disclosure, at least one of the data acquisition module 610, the data risk assessment module 620 and the interpretation module 630 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application-specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way of integrating or packaging the circuit, or implemented in any one of the three implementation methods of software, hardware and firmware or in any appropriate combination of any of them. Alternatively, at least one of the data acquisition module 610, the data risk assessment module 620 and the interpretation module 630 can be at least partially implemented as a computer program module, and when the computer program module is run, the corresponding function can be executed.

[0120] Figure 7 A block diagram of an electronic device suitable for implementing a data risk assessment method according to an embodiment of the present disclosure is schematically shown.

[0121] like Figure 7 As shown, the electronic device 700 according to an embodiment of the present disclosure includes a processor 701, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 702 or a program loaded from a storage part 708 to a random access memory (RAM) 703. The processor 701 may include, for example, a general-purpose microprocessor (such as a CPU), an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (for example, an application-specific integrated circuit (ASIC)), etc. The processor 701 may also include an onboard memory for caching purposes. The processor 701 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0122] In RAM 703, various programs and data required for the operation of electronic device 700 are stored. Processor 701, ROM 702 and RAM 703 are connected to each other via bus 704. Processor 701 performs various operations of the method flow according to the embodiment of the present disclosure by executing the program in ROM 702 and / or RAM 703. It should be noted that the program can also be stored in one or more memories other than ROM 702 and RAM 703. Processor 701 can also perform various operations of the method flow according to the embodiment of the present disclosure by executing the program stored in one or more memories.

[0123] According to an embodiment of the present disclosure, the electronic device 700 may further include an input / output (I / O) interface 705, which is also connected to the bus 704. The electronic device 700 may further include one or more of the following components connected to the input / output (I / O) interface 705: an input portion 706 including a keyboard, a mouse, etc.; an output portion 707 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage portion 708 including a hard disk, etc.; and a communication portion 709 including a network interface card such as a LAN card, a modem, etc. The communication portion 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the input / output (I / O) interface 705 as needed. A removable medium 711, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 710 as needed, so that a computer program read therefrom is installed into the storage portion 708 as needed.

[0124] The present disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or may exist independently without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present disclosure is implemented.

[0125] According to an embodiment of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, it may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, an apparatus or a device. For example, according to an embodiment of the present disclosure, the computer-readable storage medium may include the ROM 702 and / or RAM 703 described above and / or one or more memories other than ROM 702 and RAM 703.

[0126] The embodiment of the present disclosure also includes a computer program product, which includes a computer program, and the computer program contains program code for executing the method shown in the flowchart. When the computer program product is run in a computer system, the program code is used to enable the computer system to implement the data risk assessment method provided by the embodiment of the present disclosure.

[0127] The above functions defined in the system / device of the embodiment of the present disclosure are performed when the computer program is executed by the processor 701. According to the embodiment of the present disclosure, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0128] In one embodiment, the computer program may rely on tangible storage media such as optical storage devices, magnetic storage devices, etc. In another embodiment, the computer program may also be transmitted and distributed in the form of signals on a network medium, and downloaded and installed through the communication part 709, and / or installed from the removable medium 711. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0129] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 709, and / or installed from the removable medium 711. When the computer program is executed by the processor 701, the above functions defined in the system of the embodiment of the present disclosure are performed. According to the embodiment of the present disclosure, the system, device, means, module, unit, etc. described above can be implemented by a computer program module.

[0130] According to an embodiment of the present disclosure, the program code for executing the computer program provided by the embodiment of the present disclosure can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level process and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, Java, C++, python, "C" language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on the remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect through the Internet).

[0131] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0132] It will be appreciated by those skilled in the art that the features described in the various embodiments of the present disclosure may be combined and / or combined in a variety of ways, even if such combinations or combinations are not explicitly described in the present disclosure. In particular, without departing from the spirit and teachings of the present disclosure, the features described in the various embodiments of the present disclosure may be combined and / or combined in a variety of ways. All of these combinations and / or combinations fall within the scope of the present disclosure.

[0133] The embodiments of the present disclosure are described above. However, these embodiments are only for illustrative purposes and are not intended to limit the scope of the present disclosure. Although the embodiments are described above, this does not mean that the measures in the various embodiments cannot be used in combination to advantage. Without departing from the scope of the present disclosure, those skilled in the art may make a variety of substitutions and modifications, which should all fall within the scope of the present disclosure.

Claims

1. A data risk assessment method, characterized in that: The method comprises: Obtain credit data to be assessed; Based on the risk assessment model deployed on the target system, the credit data to be assessed is assessed to obtain an assessment result, wherein the risk assessment model is obtained through pre-training; The evaluation results are explained using interpretability technology to obtain explanation results, and the explanation results are visualized.

2. The data risk assessment method according to claim 1, characterized in that: The risk assessment model is obtained through pre-training, including: Get the original data; Dividing and preprocessing the original data, wherein the original data is divided into a data training set, a data test set and a data verification set; A risk assessment model is constructed, and the risk assessment model is trained using the preprocessed data training set to obtain a trained risk assessment model.

3. The data risk assessment method according to claim 1 or 2, characterized in that: The risk assessment model is obtained through pre-training and also includes: Using the preprocessed data test set to test the trained risk assessment model to obtain a predictive ability indicator of the risk assessment model; Based on the predictive ability indicator, the parameters of the risk assessment model are adjusted using the data validation set to optimize the risk assessment model.

4. The data risk assessment method according to claim 1 or 2, characterized in that: The risk assessment model is deployed on a target system, and the deployment method includes: Converting the trained risk assessment model into a target format according to the requirements of the target system; The risk assessment model converted into the target format is packaged into an image and deployed on the target system.

5. The data risk assessment method according to claim 4, characterized in that: The risk assessment model is deployed on the target system, and the deployment method further includes: Performing unit testing, integration testing and load testing on the target system to verify the performance of the target system and the accuracy of the risk assessment model in the target system; The performance of the target system and the accuracy of the risk assessment model in the target system are monitored.

6. The data risk assessment method according to claim 1, characterized in that: The interpretability technology is used to interpret the evaluation results to obtain the interpretation results, including: Obtaining the evaluation result, and defining the evaluation result as a target sample; Generate a plurality of neighborhood samples near the feature space of the target sample; Input the neighborhood samples into the original black box model to obtain the prediction result of each sample; Based on the characteristics and prediction results of the neighborhood samples, a local explanatory model is fitted using linear regression to determine the influence of each feature on the model prediction.

7. The data risk assessment method according to claim 1 or 6, characterized in that: The visualizing the interpretation result comprises: Present the interpretation results in a graphical form to show the impact of different features on the evaluation results; The interactions between the features were analyzed based on the interpretation results to identify the main influencing factors.

8. The data risk assessment method according to claim 2, characterized in that: The preprocessing of the raw data comprises: Clean the raw data; De-duplication, filling missing values, converting data types, and performing data desensitization processing on the cleaned raw data; Based on business needs, feature extraction is performed on the desensitized original data.

9. A data risk assessment device, characterized in that: The device comprises: A data acquisition module, used to acquire credit data to be evaluated; A data risk assessment module, used to assess the credit data to be assessed based on a risk assessment model deployed on a target system to obtain an assessment result, wherein the risk assessment model is obtained through pre-training; The explanation module is used to explain the evaluation results using explainability technology to obtain explanation results and visualize the explanation results.

10. The data risk assessment device according to claim 9, characterized in that: The data risk assessment module includes a model training submodule and a model optimization submodule, wherein the model training submodule includes: A raw data acquisition unit, used for acquiring raw data; A preprocessing unit, used to divide and preprocess the original data, wherein the original data is divided into a data training set, a data test set and a data verification set; The model training unit is used to construct a risk assessment model and train the risk assessment model using the preprocessed data training set to obtain a trained risk assessment model.

11. The data risk assessment device according to claim 10, characterized in that: The model optimization submodule includes: A model testing unit, used to test the trained risk assessment model using a preprocessed data test set to obtain a prediction capability index of the risk assessment model; A model adjustment unit is used to adjust the parameters of the risk assessment model based on the prediction ability indicator and use the data verification set to optimize the risk assessment model.

12. An electronic device comprising: one or more processors; a memory for storing one or more computer programs, It is characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 8.

13. A computer-readable storage medium having a computer program or instruction stored thereon, characterized in that: When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

14. A computer program product comprising a computer program or instructions, characterized in that When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images