Artificial intelligence-based data checking method, device, equipment and storage medium

Abstract

The application belongs to the field of artificial intelligence and the field of financial technology, and relates to a data verification method based on artificial intelligence, which comprises the following steps: judging whether a data verification request triggered by a user is received; if yes, obtaining a policy number and target dimension scene characteristics input by the user; obtaining maintenance characteristics corresponding to the target dimension scene characteristics based on the policy number; obtaining a specified database storage result corresponding to the maintenance characteristics based on the policy number; generating a prediction result corresponding to the maintenance characteristics through a test prediction model; and performing verification processing on the specified database storage result based on the prediction result to generate a corresponding verification result. The application also provides a data verification device based on artificial intelligence, a computer device and a storage medium. In addition, the application also relates to blockchain technology, and the verification result can be stored in a blockchain. The application can be applied to a business result verification scene in the financial field, improves the processing efficiency of data verification, and ensures the accuracy of the generated verification result corresponding to the database storage result.

Description

Technical Field

[0001] This application relates to the fields of artificial intelligence development technology and financial technology, and in particular to data verification methods, devices, computer equipment and storage media based on artificial intelligence. Background Technology

[0002] As insurance companies continue to develop their business and optimize their development capabilities, data testers within these companies also need to continuously improve their productivity and efficiency. They must identify areas for improvement based on the characteristics of their daily work to enhance data testing efficiency. For upstream business systems, many changes to requirements can impact the normal operation of downstream systems. Each change requires joint testing with downstream systems to verify their functionality. This includes a significant portion of non-impact joint testing, meaning changes to the intermediate logic of the upstream system should not affect the final data stored in the database. The data and field business logic provided to downstream systems must remain unchanged, effectively ensuring that the data usage logic of downstream systems is unaffected.

[0003] To address the business requirement of data verification in business systems, insurance companies currently primarily rely on manual processing to assess the impact and conduct inter-system integration verification. This involves expending manpower to manually query the database of unmodified downstream business systems to verify the correctness of the data. However, if upstream business systems can ensure the correctness of their data results, this manpower for integration can be saved, thereby improving the efficiency of data testing and verification. Therefore, the existing data verification methods for business systems suffer from high workload and low efficiency, necessitating a data verification method that can effectively improve the processing efficiency of data verification in business systems. Summary of the Invention

[0004] The purpose of this application is to propose a data verification method, apparatus, computer equipment, and storage medium based on artificial intelligence, so as to solve the technical problems of large processing workload and low processing efficiency in the existing data verification methods of business systems.

[0005] To address the aforementioned technical problems, this application provides a data verification method based on artificial intelligence, employing the following technical solution:

[0006] Determine whether a user-triggered data verification request has been received;

[0007] If so, obtain the policy number entered by the user, and obtain the target dimension scene features selected by the user;

[0008] Based on the policy number, obtain the maintenance characteristics corresponding to the target dimension scenario features;

[0009] Based on the policy number, obtain the specified data entry result corresponding to the maintenance feature;

[0010] The target dimension scene features and the maintenance characteristics are predicted and processed by a preset test prediction model to generate prediction results corresponding to the maintenance characteristics;

[0011] Based on the prediction results, the specified landing result is verified to generate a verification result corresponding to the specified landing result.

[0012] Furthermore, the step of obtaining the maintenance characteristics corresponding to the target dimension scenario features based on the policy number specifically includes:

[0013] Retrieve the application form table corresponding to the policy number;

[0014] Based on the target dimension scenario features, the insurance application form is processed by data query, and specified features that match the target dimension scenario features are selected from the insurance application form.

[0015] Obtain the feature parameters corresponding to the specified feature from the insurance application form;

[0016] The aforementioned feature parameters are used as maintenance features corresponding to the target dimension scene features.

[0017] Furthermore, the step of obtaining the specified database entry result corresponding to the maintenance feature based on the policy number specifically includes:

[0018] Call the policy table corresponding to the policy number;

[0019] Based on the maintenance characteristics, the policy table is queried and processed to filter out the specified characteristics that match the maintenance characteristics from the policy table;

[0020] Retrieve the characteristic result value corresponding to the specified characteristic from the policy table;

[0021] The characteristic result value is used as the specified database result corresponding to the maintenance characteristic.

[0022] Furthermore, the step of verifying the specified landing result based on the prediction result and generating a verification result corresponding to the specified landing result specifically includes:

[0023] Determine whether the predicted result is the same as the specified landing result;

[0024] If the prediction result is the same as the specified landing result, then a first verification result is generated to show that the specified landing result has passed the verification.

[0025] If the prediction result is different from the specified result, a second verification result is generated indicating that the specified result failed the verification.

[0026] Furthermore, before the step of performing prediction processing on the target dimension scene features and the maintenance characteristics using a preset test prediction model to generate a prediction result corresponding to the maintenance characteristics, the method further includes:

[0027] Obtain preset historical policy data, and perform data cleaning on the historical policy data to obtain target policy data;

[0028] Training data is constructed based on specified policy data; wherein, the training data includes the features of historical policies, the characteristics of historical policies, and the result set corresponding to the features of historical policies;

[0029] The preset initial model is trained based on the training data to obtain the first trained model;

[0030] Determine whether there are any abnormal results in the model prediction results output by the first model during the training process;

[0031] If so, receive the pre-set review result of the abnormal result input by the business personnel, and update the result set in the training data based on the review result and the abnormal result to obtain the updated specified training data;

[0032] The first model is retrained based on the specified training data to obtain the test prediction model that meets the preset expected conditions.

[0033] Furthermore, the step of retraining the first model based on the specified training data to obtain the test prediction model that meets the preset expected conditions specifically includes:

[0034] The first model is retrained based on the specified training data to obtain the retrained second model.

[0035] Construct test data based on the specified training data;

[0036] The second model is tested using the test data to obtain the model evaluation index corresponding to the second model;

[0037] Determine whether the model evaluation index is greater than a preset evaluation index threshold;

[0038] If so, the second model shall be used as the test prediction model.

[0039] Furthermore, after the step of verifying the specified landing result based on the prediction result and generating a verification result corresponding to the specified landing result, the method further includes:

[0040] Based on the verification results, a data analysis report corresponding to the maintenance features is generated;

[0041] Obtain the communication information of preset testers;

[0042] Based on the communication information, the data analysis report is sent to the testers.

[0043] To address the aforementioned technical problems, this application also provides an artificial intelligence-based data verification device, employing the following technical solution:

[0044] The first judgment module is used to determine whether a data verification request triggered by the user has been received;

[0045] The first acquisition module is used to acquire the policy number input by the user and the target dimension scene features selected by the user if the condition is met.

[0046] The second acquisition module is used to acquire maintenance characteristics corresponding to the target dimension scenario features based on the policy number;

[0047] The third acquisition module is used to acquire the specified database entry result corresponding to the maintenance feature based on the policy number;

[0048] The prediction module is used to perform prediction processing on the target dimension scene features and the maintenance characteristics through a preset test prediction model, and generate prediction results corresponding to the maintenance characteristics.

[0049] The first generation module is used to perform verification processing on the specified landing result based on the prediction result, and generate a verification result corresponding to the specified landing result.

[0050] To address the aforementioned technical problems, this application also provides a computer device that employs the following technical solution:

[0051] Determine whether a user-triggered data verification request has been received;

[0052] If so, obtain the policy number entered by the user, and obtain the target dimension scene features selected by the user;

[0053] Based on the policy number, obtain the maintenance characteristics corresponding to the target dimension scenario features;

[0054] Based on the policy number, obtain the specified data entry result corresponding to the maintenance feature;

[0055] The target dimension scene features and the maintenance characteristics are predicted and processed by a preset test prediction model to generate prediction results corresponding to the maintenance characteristics;

[0056] Based on the prediction results, the specified landing result is verified to generate a verification result corresponding to the specified landing result.

[0057] To address the aforementioned technical problems, this application also provides a computer-readable storage medium, employing the technical solution described below:

[0058] Determine whether a user-triggered data verification request has been received;

[0059] If so, obtain the policy number entered by the user, and obtain the target dimension scene features selected by the user;

[0060] Based on the policy number, obtain the maintenance characteristics corresponding to the target dimension scenario features;

[0061] Based on the policy number, obtain the specified data entry result corresponding to the maintenance feature;

[0062] The target dimension scene features and the maintenance characteristics are predicted and processed by a preset test prediction model to generate prediction results corresponding to the maintenance characteristics;

[0063] Based on the prediction results, the specified landing result is verified to generate a verification result corresponding to the specified landing result.

[0064] Compared with the prior art, the embodiments of this application have the following main advantages:

[0065] This embodiment first determines whether a user-triggered data verification request has been received. If so, it obtains the policy number input by the user and the target dimension scene feature selected by the user. Then, based on the policy number, it obtains the maintenance feature corresponding to the target dimension scene feature. Next, based on the policy number, it obtains the specified database entry result corresponding to the maintenance feature. Subsequently, it uses a preset test prediction model to perform prediction processing on the target dimension scene feature and the maintenance feature, generating a prediction result corresponding to the maintenance feature. Finally, it performs verification processing on the specified database entry result based on the prediction result, generating a verification result corresponding to the specified database entry result. In this embodiment, when the user-triggered data verification request is received, and the user inputs the policy number and the selected target dimension scene feature, the embodiment first obtains the maintenance feature corresponding to the target dimension scene feature based on the policy number, and then obtains the specified database entry result corresponding to the maintenance feature based on the policy number. Then, it uses a preset test prediction model to perform prediction processing on the target dimension scene feature and the maintenance feature, generating a prediction result corresponding to the maintenance feature. Subsequently, it performs verification processing on the specified database entry result based on the prediction result, thereby achieving fast and accurate generation of a verification result corresponding to the specified database entry result. This application provides a fast and accurate data verification method based on the use of a test prediction model. It eliminates the need for manual database queries to verify the correctness of the data insertion results, effectively improving the processing efficiency of data verification and ensuring the accuracy of the generated verification results corresponding to the data insertion results. Attached Figure Description

[0066] To more clearly illustrate the solutions in this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0067] Figure 1 This is an exemplary system architecture diagram to which this application can be applied;

[0068] Figure 2 A flowchart of an embodiment of the AI-based data verification method according to this application;

[0069] Figure 3 This is a schematic diagram of a structure of an embodiment of the AI-based data verification device according to this application;

[0070] Figure 4 This is a schematic diagram of the structure of one embodiment of the computer device according to this application. Detailed Implementation

[0071] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings of this application, are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings of this application are used to distinguish different objects, not to describe a particular order.

[0072] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0073] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0074] like Figure 1 As shown, system architecture 100 may include terminal devices 101, 102, and 103, a network 104, and a server 105. Network 104 serves as the medium for providing communication links between terminal devices 101, 102, and 103 and server 105. Network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0075] Users can use terminal devices 101, 102, and 103 to interact with server 105 via network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, and 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social media platform software, etc.

[0076] Terminal devices 101, 102, and 103 can be various electronic devices with displays and support web browsing, including but not limited to smartphones, tablets, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III), MP4 players (Moving Picture Experts Group Audio Layer IV), laptops, and desktop computers, etc.

[0077] Server 105 can be a server that provides various services, such as a backend server that supports the pages displayed on terminal devices 101, 102, and 103.

[0078] It should be noted that the AI-based data verification method provided in this application is generally executed by a server / terminal device, and correspondingly, the AI-based data verification device is generally located in the server / terminal device.

[0079] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0080] Continue to refer to Figure 2 This document illustrates a flowchart of an embodiment of the AI-based data verification method according to this application. The order of steps in the flowchart can be changed, and some steps can be omitted, depending on different requirements. The AI-based data verification method provided in this application can be applied to any scenario requiring business data verification, and thus can be applied to products in these scenarios, such as data verification of backlogged business data in the financial insurance field. The AI-based data verification method includes the following steps:

[0081] Step S201: Determine whether a data verification request triggered by the user has been received.

[0082] In this embodiment, the AI-based data verification method operates on an electronic device (e.g., Figure 1 The server / terminal device shown can obtain data verification requests via wired or wireless connections. It should be noted that the aforementioned wireless connection methods may include, but are not limited to, 3G / 4G / 5G connections, WiFi connections, Bluetooth connections, WiMAX connections, Zigbee connections, UWB (ultra-wideband) connections, and other currently known or future-developed wireless connection methods. The specific execution entity of the aforementioned AI-based data verification method can be an upstream business system applied within an insurance company, such as the headquarters insurance business system. In the business scenario of data verification for financial insurance data, the headquarters insurance business system may include the insurance company's headquarters auto insurance business system, life insurance business system, claims business system, etc. Within the system business architecture, insurance companies also develop downstream business systems corresponding to the upstream business systems, such as branch auto insurance business systems, branch insurance company headquarters life insurance business systems, branch insurance company headquarters claims business systems, etc.

[0083] Step S202: If yes, obtain the policy number entered by the user and obtain the target dimension scene features selected by the user.

[0084] In this embodiment, after a user triggers a data verification request in the upstream business system, they can further enter the policy number on the verification page of the upstream business system. The verification page will pre-provide multiple dimensional scenario features for the user to select. The target dimensional scenario feature selected by the user can be determined by acquiring the user's selection operation for the dimensional scenario features. In the business scenario of data verification for financial insurance data, the aforementioned dimensional scenario features can at least include scenario types such as maintenance entrusted funds, long-term critical illness insurance, and short-term insurance; product dimensions such as inclusive insurance and family policies; and features such as intermediaries, policy source, payment method, fee type, coverage mode, and co-insurance type.

[0085] Step S203: Based on the policy number, obtain the maintenance characteristics corresponding to the target dimension scenario features.

[0086] In this embodiment, the specific implementation process of obtaining the maintenance characteristics corresponding to the target dimension scenario features based on the policy number will be further described in detail in subsequent specific embodiments of this application, and will not be elaborated on here.

[0087] Step S204: Based on the policy number, obtain the specified data entry result corresponding to the maintenance feature.

[0088] In this embodiment, the specific implementation process of obtaining the specified data entry result corresponding to the maintenance feature based on the policy number will be further described in detail in subsequent specific embodiments of this application, and will not be elaborated on here.

[0089] Step S205: The target dimension scene features and the maintenance characteristics are predicted using a preset test prediction model to generate prediction results corresponding to the maintenance characteristics.

[0090] In this embodiment, the target dimension scene features and the maintenance characteristics can be input into the test prediction model to perform prediction processing on the target dimension scene features and the maintenance characteristics, and then output a prediction result corresponding to the maintenance characteristics. The model construction process of the above-mentioned test prediction model will be described in more detail in subsequent specific embodiments of this application, and will not be elaborated upon here.

[0091] Step S206: Based on the prediction result, perform verification processing on the specified landing result to generate a verification result corresponding to the specified landing result.

[0092] In this embodiment, the specific implementation process of verifying the specified landing result based on the prediction result and generating a verification result corresponding to the specified landing result will be further described in detail in subsequent specific embodiments of this application, and will not be elaborated on here.

[0093] This application first determines whether a user-triggered data verification request has been received. If so, it obtains the policy number input by the user and the target dimension scene feature selected by the user. Then, based on the policy number, it obtains the maintenance feature corresponding to the target dimension scene feature. Next, based on the policy number, it obtains the specified database entry result corresponding to the maintenance feature. Subsequently, it uses a preset test prediction model to perform prediction processing on the target dimension scene feature and the maintenance feature to generate a prediction result corresponding to the maintenance feature. Finally, it performs verification processing on the specified database entry result based on the prediction result to generate a verification result corresponding to the specified database entry result. When this application obtains the policy number and selected target dimension scene feature input by the user after triggering a data verification request, it first obtains the maintenance feature corresponding to the target dimension scene feature based on the policy number, and then obtains the specified database entry result corresponding to the maintenance feature based on the policy number. Then, it uses a preset test prediction model to perform prediction processing on the target dimension scene feature and the maintenance feature to generate a prediction result corresponding to the maintenance feature. Subsequently, it performs verification processing on the specified database entry result based on the prediction result to achieve fast and accurate generation of a verification result corresponding to the specified database entry result. This application provides a fast and accurate data verification method based on the use of a test prediction model. It eliminates the need for manual database queries to verify the correctness of the data insertion results, effectively improving the processing efficiency of data verification and ensuring the accuracy of the generated verification results corresponding to the data insertion results.

[0094] In some alternative implementations, step S203 includes the following steps:

[0095] Call the application form table corresponding to the policy number.

[0096] In this embodiment, different dimensional scenario features and insurance application forms are pre-established according to actual business needs. The insurance application forms are used to obtain feature and characteristic input parameters.

[0097] Based on the target dimension scenario features, the insurance application form is processed by data query, and specified features that match the target dimension scenario features are selected from the insurance application form.

[0098] In this embodiment, the specified feature that matches the target dimension scene feature specifically refers to the feature stored in the insurance application form that is the same as the target dimension scene feature.

[0099] Obtain the feature parameters corresponding to the specified feature from the insurance application form.

[0100] In this embodiment, after determining the specified feature that matches the target dimension scene feature, the feature characteristic input parameters corresponding to the specified feature can be further queried from the insurance application table.

[0101] The aforementioned feature parameters are used as maintenance features corresponding to the target dimension scene features.

[0102] This application calls the application form table corresponding to the policy number; then, based on the target dimension scenario features, it performs data query processing on the application form table, filtering out specified features that match the target dimension scenario features from the application form table; subsequently, it obtains the feature characteristic input parameters corresponding to the specified features from the application form table; and finally, it uses the feature characteristic input parameters as maintenance characteristics corresponding to the target dimension scenario features. This application, by calling the application form table corresponding to the policy number and then performing data query processing on the target dimension scenario features based on the use of the application form table, can quickly and accurately obtain the maintenance characteristics corresponding to the target dimension scenario features from the application form table, improving the efficiency and intelligence of maintenance characteristic acquisition.

[0103] In some optional implementations of this embodiment, step S204 includes the following steps:

[0104] Call the policy table corresponding to the policy number.

[0105] In this embodiment, different mapping relationships between features and policy tables are pre-established according to actual business needs. The policy tables are used to obtain feature result values.

[0106] Based on the maintenance characteristics, the policy table is queried and processed to select the specified characteristics that match the maintenance characteristics from the policy table.

[0107] In this embodiment, the specified characteristic that matches the maintenance characteristic specifically refers to the characteristic stored in the policy table that is the same as the maintenance characteristic.

[0108] Retrieve the characteristic result value corresponding to the specified characteristic from the policy table.

[0109] In this embodiment, after determining the specified characteristic that matches the maintenance characteristic, the characteristic result value corresponding to the specified characteristic can be further queried from the policy table.

[0110] The characteristic result value is used as the specified database result corresponding to the maintenance characteristic.

[0111] This application retrieves the specified database entry for the policy number by calling the policy table; then, based on the maintenance characteristics, it performs data query processing on the policy table to filter out the specified characteristics that match the maintenance characteristics; subsequently, it obtains the characteristic result value corresponding to the specified characteristics from the policy table; and finally, it uses the characteristic result value as the specified database entry result corresponding to the maintenance characteristics. This application, by calling the policy table corresponding to the policy number and then performing data query processing on the maintenance characteristics based on the use of the policy table, can quickly and accurately obtain the specified database entry result corresponding to the maintenance characteristics from the policy table, improving the efficiency and intelligence of obtaining the specified database entry result.

[0112] In some alternative implementations, step S206 includes the following steps:

[0113] Determine whether the predicted result is the same as the specified landing result.

[0114] In this embodiment, the prediction result and the specified data location result can be compared to determine whether the prediction result and the specified data location result are the same data.

[0115] If the prediction result is the same as the specified result, then a first verification result is generated to show that the specified result has passed the verification.

[0116] In this embodiment, if the prediction result is the same as the specified database result, the specified database result will be determined to be accurate result data, and a first verification result will be generated to show that the specified database result has passed the verification.

[0117] If the prediction result is different from the specified result, a second verification result is generated indicating that the specified result failed the verification.

[0118] In this embodiment, if the prediction result is different from the specified database result, the specified database result will be determined to be erroneous, and a first verification result will be generated to show that the specified database result has passed the verification.

[0119] This application determines whether the predicted result is the same as the specified database result. If the predicted result is the same as the specified database result, a first verification result is generated indicating that the specified database result has passed verification. If the predicted result is different from the specified database result, a second verification result is generated indicating that the specified database result has failed verification. By comparing the content of the predicted result and the specified database result, this application can quickly and accurately generate a verification result corresponding to the specified database result, thus ensuring the accuracy of the generated verification result for the specified database result.

[0120] In some alternative implementations, prior to step S205, the electronic device may also perform the following steps:

[0121] Obtain preset historical policy data, and perform data cleaning on the historical policy data to obtain target policy data.

[0122] In this embodiment, the aforementioned historical policy data can be policy data collected within a historical time period. This historical policy data includes at least the characteristics of historical policies, the features of historical policies, and the result set corresponding to the characteristics of historical policies, and may also include other business data. The value of the historical time period is not specifically limited and can be set according to actual business needs. Furthermore, the aforementioned data cleaning may include cleaning outliers and dirty data.

[0123] Training data is constructed based on specified policy data.

[0124] In this embodiment, the training data includes the features of historical policies, the characteristics of historical policies, and the result set corresponding to the features of historical policies.

[0125] The preset initial model is trained based on the training data to obtain the first trained model.

[0126] In this embodiment, the initial model can specifically be a classification model, such as any one of the following: logistic regression, decision tree, support vector machine, or Naive Bayes. The training process for the initial model can refer to the training process of commonly used classification models, and will not be elaborated upon here.

[0127] Determine whether there are any abnormal results in the model prediction results output by the first model during the training process.

[0128] In this embodiment, the model prediction result output by the first model during training, which corresponds to the characteristics of a specified historical policy, is compared with the result set corresponding to that specified historical policy. If the two sets of data are not identical, the model prediction result is determined to be an abnormal result. Here, the specified historical policy is any one of all historical policies. For example, if the specified historical policy's characteristic is a payment status feature, and the feature of the payment status feature is the payment status in the policy table, and the result value of the payment status is "fully paid," but the actual data entry result corresponding to the payment status of the specified historical policy's payment status is "partially paid," a discrepancy result will be indicated, meaning that the model prediction result output by the first model is different from the corresponding actual data entry result.

[0129] If so, the system receives the audit results of the abnormal results input by the pre-set business personnel, and updates the result set in the training data based on the audit results and the abnormal results to obtain the updated specified training data.

[0130] In this embodiment, the business personnel can be testers of the business system. The aforementioned review results can include whether the model prediction result is normal or incorrect. The process of updating the result set in the training data based on the review results and the incorrect results to obtain updated specified training data can include: if the model prediction result is normal, no changes are made to the result set corresponding to that model prediction result; if the model prediction result is incorrect, the corresponding actual data is used to modify the result set corresponding to that model prediction result. Alternatively, if some business data is missing from the result set corresponding to the model prediction result, the result set corresponding to the model prediction result is populated with the missing business data to correct the result set. Based on the above operations, the result set in the training data is updated, and the updated specified training data is obtained.

[0131] The first model is retrained based on the specified training data to obtain the test prediction model that meets the preset expected conditions.

[0132] In this embodiment, the specific implementation process of retraining the first model based on the specified training data to obtain the test prediction model that meets the preset expected conditions will be further described in detail in subsequent specific embodiments of this application, and will not be elaborated on here.

[0133] This application obtains target policy data by acquiring preset historical policy data and cleaning the historical policy data; then, it constructs training data based on the specified policy data; subsequently, it trains a preset initial model based on the training data to obtain a trained first model; next, it determines whether there are abnormal results in the model prediction results output by the first model during training; if so, it receives the review results of the abnormal results input by preset business personnel, and updates the result set in the training data based on the review results and the abnormal results to obtain updated specified training data; finally, it retrains the first model based on the specified training data to obtain the test prediction model that meets preset expected conditions. This application constructs training data using preset historical policy data, and then trains a preset initial model based on the training data to obtain a trained first model. During the training process, if abnormal results are detected in the model prediction results output by the first model, it intelligently receives the review results of the abnormal results input by preset business personnel, and updates the result set in the training data based on the review results and the abnormal results to obtain updated specified training data. Subsequently, the first model is retrained using the updated specified training data to obtain the final test prediction model that meets the expected conditions. This effectively ensures the model effect and prediction accuracy of the generated test prediction model and improves the construction efficiency of the test prediction model.

[0134] In some optional implementations of this embodiment, retraining the first model based on the specified training data to obtain the test prediction model that meets preset expected conditions includes the following steps:

[0135] The first model is retrained based on the specified training data to obtain the retrained second model.

[0136] In this embodiment, the process of retraining the first model based on the specified training data can refer to the process of training the preset initial model based on the training data to obtain the trained first model, which will not be elaborated on here.

[0137] Test data is constructed based on the specified training data.

[0138] In this embodiment, data corresponding to the preset proportion can be randomly selected from the specified training data to obtain the corresponding test data.

[0139] The second model is tested using the test data to obtain the model evaluation index corresponding to the second model.

[0140] In this embodiment, the second model can be tested using the test data to calculate the model evaluation index corresponding to the second model. The selection of the model evaluation index is not specifically limited and can be determined according to the actual model construction needs. For example, it may include one or more of the following indices: accuracy, F1 score, recall, etc., of the prediction results obtained after the second model performs result prediction processing on the test data.

[0141] Determine whether the model evaluation index is greater than a preset evaluation index threshold.

[0142] In this embodiment, the values ​​of the above-mentioned evaluation index thresholds are not specifically limited and can be set according to actual usage requirements.

[0143] If so, the second model shall be used as the test prediction model.

[0144] This application retrains the first model based on the specified training data to obtain a retrained second model; then, test data is constructed based on the specified training data; subsequently, the second model is tested using the test data to obtain a model evaluation index corresponding to the second model; next, it is determined whether the model evaluation index is greater than a preset evaluation index threshold; if so, the second model is used as the test prediction model. This application retrains the first model using the specified training data to obtain a retrained second model, and further tests the second model using test data constructed from the specified training data. The second model that passes the test, i.e., whose model evaluation index is greater than the evaluation index threshold, is used as the final test prediction model to complete the test prediction model construction process. This effectively ensures the model performance and prediction accuracy of the generated test prediction model and improves the efficiency of test prediction model construction.

[0145] In some optional implementations of this embodiment, after step S206, the electronic device may further perform the following steps:

[0146] A data analysis report corresponding to the maintenance feature is generated based on the verification results.

[0147] In this embodiment, a data analysis report corresponding to the maintenance features can be generated by filling the aforementioned maintenance features and the verification result into the corresponding positions within a preset report template. The content of the report template is not specifically limited and can be set according to actual business needs.

[0148] Obtain the communication information of the pre-set testers.

[0149] In this embodiment, the testers mentioned above can be internal testers who perform data processing tests for the business system. The communication information mentioned above may include email addresses or telephone numbers.

[0150] Based on the communication information, the data analysis report is sent to the testers.

[0151] In this embodiment, the data analysis report can be sent to the communication terminal corresponding to the tester based on communication information.

[0152] This application generates a data analysis report corresponding to the maintenance feature based on the verification result; then obtains the communication information of preset testers; and subsequently sends the data analysis report to the testers based on the communication information. After generating the verification result corresponding to the specified data entry result, this application will also intelligently generate a data analysis report corresponding to the maintenance feature based on the verification result, and send the data analysis report to the corresponding testers, so that the testers can quickly understand the content of the required data verification result by reviewing the data analysis report, thereby improving the user experience for testers.

[0153] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0154] It should be emphasized that, to further ensure the privacy and security of the above verification results, the verification results can also be stored in a node of a blockchain.

[0155] The blockchain referred to in this application is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. Essentially, a blockchain is a decentralized database, a chain of data blocks linked together using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and generate the next block. A blockchain can include an underlying blockchain platform, a platform product service layer, and an application service layer.

[0156] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0157] Foundational technologies for artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0158] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing related hardware with computer-readable instructions. These computer-readable instructions can be stored in a computer-readable storage medium. When executed, the program can include the processes of the embodiments of the above methods. The aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, optical disk, or read-only memory (ROM), or random access memory (RAM).

[0159] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0160] Further reference Figure 3 As a response to the above Figure 2 To implement the method shown, this application provides an embodiment of an artificial intelligence-based data verification device, which is similar to... Figure 2 Corresponding to the method embodiments shown, this device can be specifically applied to various electronic devices.

[0161] like Figure 3 As shown, the artificial intelligence-based data verification device 300 described in this embodiment includes: a first judgment module 301, a first acquisition module 302, a second acquisition module 303, a third acquisition module 304, a prediction module 305, and a first generation module 306. Wherein:

[0162] The first judgment module 301 is used to determine whether a data verification request triggered by the user has been received.

[0163] The first acquisition module 302 is used to acquire the policy number input by the user and the target dimension scene features selected by the user if the condition is met.

[0164] The second acquisition module 303 is used to acquire maintenance characteristics corresponding to the target dimension scene features based on the policy number;

[0165] The third acquisition module 304 is used to acquire a specified database entry result corresponding to the maintenance feature based on the policy number;

[0166] The prediction module 305 is used to perform prediction processing on the target dimension scene features and the maintenance characteristics through a preset test prediction model, and generate a prediction result corresponding to the maintenance characteristics.

[0167] The first generation module 306 is used to perform verification processing on the specified landing result based on the prediction result, and generate a verification result corresponding to the specified landing result.

[0168] In this embodiment, the operations performed by the above modules or units correspond one-to-one with the steps of the artificial intelligence-based data verification method in the aforementioned implementation method, and will not be repeated here.

[0169] In some optional implementations of this embodiment, the second acquisition module 303 includes:

[0170] The first calling submodule is used to call the application form table corresponding to the policy number;

[0171] The first filtering submodule is used to perform data query processing on the insurance application form based on the target dimension scene features, and to filter out the specified features that match the target dimension scene features from the insurance application form;

[0172] The first acquisition submodule is used to obtain the feature characteristic input parameters corresponding to the specified feature from the insurance application form table;

[0173] The first determining submodule is used to take the feature parameter as a maintenance feature corresponding to the target dimension scene feature.

[0174] In this embodiment, the operations performed by the above modules or units correspond one-to-one with the steps of the artificial intelligence-based data verification method in the aforementioned implementation method, and will not be repeated here.

[0175] In some optional implementations of this embodiment, the third acquisition module 304 includes:

[0176] The second calling submodule is used to call the policy table corresponding to the policy number;

[0177] The second filtering submodule is used to perform data query processing on the policy table based on the maintenance characteristics, and to filter out the specified characteristics that match the maintenance characteristics from the policy table;

[0178] The second acquisition submodule is used to acquire the characteristic result value corresponding to the specified characteristic from the policy table;

[0179] The second determining submodule is used to use the characteristic result value as the specified database result corresponding to the maintenance characteristic.

[0180] In this embodiment, the operations performed by the above modules or units correspond one-to-one with the steps of the artificial intelligence-based data verification method in the aforementioned implementation method, and will not be repeated here.

[0181] In some optional implementations of this embodiment, the first generation module 306 includes:

[0182] The first judgment submodule is used to determine whether the prediction result is the same as the specified landing result;

[0183] The first generation submodule is used to generate a first verification result if the prediction result is the same as the specified landing result;

[0184] The second generation submodule is used to generate a second verification result if the prediction result is different from the specified location result, in which case the specified location result fails the verification.

[0185] In this embodiment, the operations performed by the above modules or units correspond one-to-one with the steps of the artificial intelligence-based data verification method in the aforementioned implementation method, and will not be repeated here.

[0186] In some optional implementations of this embodiment, the AI-based data verification device further includes:

[0187] The fourth acquisition module is used to acquire preset historical policy data and perform data cleaning on the historical policy data to obtain target policy data.

[0188] A construction module is used to construct training data based on specified policy data; wherein, the training data includes features of historical policies, characteristics of historical policies, and result sets corresponding to the features of historical policies;

[0189] The first training module is used to train a preset initial model based on the training data to obtain a trained first model.

[0190] The second judgment module is used to determine whether there are any abnormal results in the model prediction results output by the first model during the training process;

[0191] The update module is used to receive the review result of the abnormal result input by the preset business personnel if the result is true, and update the result set in the training data based on the review result and the abnormal result to obtain the updated specified training data.

[0192] The second training module is used to retrain the first model based on the specified training data to obtain the test prediction model that meets the preset expected conditions.

[0193] In this embodiment, the operations performed by the above modules or units correspond one-to-one with the steps of the artificial intelligence-based data verification method in the aforementioned implementation method, and will not be repeated here.

[0194] In some optional implementations of this embodiment, the second training module includes:

[0195] The training submodule is used to retrain the first model based on the specified training data to obtain a retrained second model.

[0196] A submodule is constructed to build test data based on the specified training data;

[0197] The testing submodule is used to test the second model using the test data to obtain the model evaluation index corresponding to the second model;

[0198] The second judgment submodule is used to determine whether the model evaluation index is greater than the preset evaluation index threshold.

[0199] The third determining submodule is used to, if so, use the second model as the test prediction model.

[0200] In this embodiment, the operations performed by the above modules or units correspond one-to-one with the steps of the artificial intelligence-based data verification method in the aforementioned implementation method, and will not be repeated here.

[0201] In some optional implementations of this embodiment, the AI-based data verification device further includes:

[0202] The second generation module is used to generate a data analysis report corresponding to the maintenance feature based on the verification result;

[0203] The fifth acquisition module is used to acquire the communication information of preset testers;

[0204] The sending module is used to send the data analysis report to the testers based on communication information.

[0205] In this embodiment, the operations performed by the above modules or units correspond one-to-one with the steps of the artificial intelligence-based data verification method in the aforementioned implementation method, and will not be repeated here.

[0206] To address the aforementioned technical problems, embodiments of this application also provide a computer device. Please refer to [link / reference needed]. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0207] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected via a system bus. It should be noted that only the computer device 4 with components 41-43 is shown in the figure; however, it should be understood that it is not required to implement all the shown components, and more or fewer components can be implemented alternatively. Those skilled in the art will understand that the computer device described here is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0208] The computer device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The computer device can interact with the user via a keyboard, mouse, remote control, touchpad, or voice control.

[0209] The memory 41 includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 41 may be an internal storage unit of the computer device 4, such as the hard disk or memory of the computer device 4. In other embodiments, the memory 41 may also be an external storage device of the computer device 4, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the computer device 4. Of course, the memory 41 may also include both the internal storage unit and its external storage device of the computer device 4. In this embodiment, the memory 41 is typically used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions based on artificial intelligence data verification methods. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or will be output.

[0210] In some embodiments, the processor 42 may be a central processing unit (CPU), controller, microcontroller, microprocessor, or other data processing chip. The processor 42 is typically used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to execute computer-readable instructions stored in the memory 41 or to process data, for example, to execute computer-readable instructions of the artificial intelligence-based data verification method.

[0211] The network interface 43 may include a wireless network interface or a wired network interface, which is typically used to establish communication connections between the computer device 4 and other electronic devices.

[0212] Compared with the prior art, the embodiments of this application have the following main advantages:

[0213] In this embodiment, it is first determined whether a data verification request triggered by a user has been received. If so, the policy number input by the user is obtained, and the target dimension scene feature selected by the user is also obtained. Then, based on the policy number, the maintenance feature corresponding to the target dimension scene feature is obtained. Next, based on the policy number, the specified database entry result corresponding to the maintenance feature is obtained. Subsequently, a preset test prediction model is used to predict the target dimension scene feature and the maintenance feature to generate a prediction result corresponding to the maintenance feature. Finally, based on the prediction result, the specified database entry result is verified to generate a verification result corresponding to the specified database entry result. In this embodiment, when the policy number and the selected target dimension scene feature input by the user after triggering a data verification request are obtained, the maintenance feature corresponding to the target dimension scene feature is first obtained based on the policy number, and the specified database entry result corresponding to the maintenance feature is obtained based on the policy number. Then, a preset test prediction model is used to predict the target dimension scene feature and the maintenance feature to generate a prediction result corresponding to the maintenance feature. Subsequently, the specified database entry result is verified based on the prediction result to achieve fast and accurate generation of a verification result corresponding to the specified database entry result. This application provides a fast and accurate data verification method based on the use of a test prediction model. It eliminates the need for manual database queries to verify the correctness of the data insertion results, effectively improving the processing efficiency of data verification and ensuring the accuracy of the generated verification results corresponding to the data insertion results.

[0214] This application also provides another embodiment, namely, providing a computer-readable storage medium storing computer-readable instructions that can be executed by at least one processor to cause the at least one processor to perform the steps of the artificial intelligence-based data verification method described above.

[0215] Compared with the prior art, the embodiments of this application have the following main advantages:

[0216] In this embodiment, it is first determined whether a data verification request triggered by a user has been received. If so, the policy number input by the user is obtained, and the target dimension scene feature selected by the user is also obtained. Then, based on the policy number, the maintenance feature corresponding to the target dimension scene feature is obtained. Next, based on the policy number, the specified database entry result corresponding to the maintenance feature is obtained. Subsequently, a preset test prediction model is used to predict the target dimension scene feature and the maintenance feature to generate a prediction result corresponding to the maintenance feature. Finally, based on the prediction result, the specified database entry result is verified to generate a verification result corresponding to the specified database entry result. In this embodiment, when the policy number and the selected target dimension scene feature input by the user after triggering a data verification request are obtained, the maintenance feature corresponding to the target dimension scene feature is first obtained based on the policy number, and the specified database entry result corresponding to the maintenance feature is obtained based on the policy number. Then, a preset test prediction model is used to predict the target dimension scene feature and the maintenance feature to generate a prediction result corresponding to the maintenance feature. Subsequently, the specified database entry result is verified based on the prediction result to achieve fast and accurate generation of a verification result corresponding to the specified database entry result. This application provides a fast and accurate data verification method based on the use of a test prediction model. It eliminates the need for manual database queries to verify the correctness of the data insertion results, effectively improving the processing efficiency of data verification and ensuring the accuracy of the generated verification results corresponding to the data insertion results.

[0217] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0218] Obviously, the embodiments described above are only some embodiments of this application, not all embodiments. The accompanying drawings show preferred embodiments of this application, but do not limit the patent scope of this application. This application can be implemented in many different forms; rather, the purpose of providing these embodiments is to provide a more thorough and comprehensive understanding of the disclosure of this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this application's specification and drawings, directly or indirectly applied to other related technical fields, are similarly within the scope of patent protection of this application.

Claims

1. A data verification method based on artificial intelligence, characterized in that, Includes the following steps: Determine whether a user-triggered data verification request has been received; If so, obtain the policy number entered by the user, and obtain the target dimension scene features selected by the user; Based on the policy number, obtain the maintenance characteristics corresponding to the target dimension scenario features; Based on the policy number, obtain the specified data entry result corresponding to the maintenance feature; The target dimension scene features and the maintenance characteristics are predicted and processed by a preset test prediction model to generate prediction results corresponding to the maintenance characteristics; Based on the prediction results, the specified landing result is verified to generate a verification result corresponding to the specified landing result. The step of obtaining the specified database entry result corresponding to the maintenance feature based on the policy number specifically includes: Call the policy table corresponding to the policy number; Based on the maintenance characteristics, the policy table is queried and processed to filter out the specified characteristics that match the maintenance characteristics from the policy table; Retrieve the characteristic result value corresponding to the specified characteristic from the policy table; The characteristic result value is used as the specified database entry result corresponding to the maintenance characteristic; In this system, a mapping relationship between different features and a policy table is pre-established, and the policy table is used to obtain the result value of the features; the specified feature that matches the maintenance feature mentioned above specifically refers to the feature stored in the policy table that is the same as the maintenance feature.

2. The data verification method based on artificial intelligence according to claim 1, characterized in that, The step of obtaining the maintenance characteristics corresponding to the target dimension scenario features based on the policy number specifically includes: Retrieve the application form table corresponding to the policy number; Based on the target dimension scenario features, the insurance application form is processed by data query, and specified features that match the target dimension scenario features are selected from the insurance application form. Obtain the feature parameters corresponding to the specified feature from the insurance application form; The aforementioned feature parameters are used as maintenance features corresponding to the target dimension scene features.

3. The data verification method based on artificial intelligence according to claim 1, characterized in that, The step of verifying the specified landing result based on the prediction result and generating a verification result corresponding to the specified landing result specifically includes: Determine whether the predicted result is the same as the specified landing result; If the prediction result is the same as the specified landing result, then a first verification result is generated to show that the specified landing result has passed the verification. If the prediction result is different from the specified result, a second verification result is generated indicating that the specified result failed the verification.

4. The data verification method based on artificial intelligence according to claim 1, characterized in that, Before the step of performing prediction processing on the target dimension scene features and the maintenance characteristics using a preset test prediction model to generate a prediction result corresponding to the maintenance characteristics, the method further includes: Obtain preset historical policy data, and perform data cleaning on the historical policy data to obtain target policy data; Training data is constructed based on specified policy data; wherein, the training data includes the features of historical policies, the characteristics of historical policies, and the result set corresponding to the features of historical policies; The preset initial model is trained based on the training data to obtain the first trained model; Determine whether there are any abnormal results in the model prediction results output by the first model during the training process; If so, receive the pre-set review result of the abnormal result input by the business personnel, and update the result set in the training data based on the review result and the abnormal result to obtain the updated specified training data; The first model is retrained based on the specified training data to obtain the test prediction model that meets the preset expected conditions.

5. The data verification method based on artificial intelligence according to claim 4, characterized in that, The step of retraining the first model based on the specified training data to obtain the test prediction model that meets the preset expected conditions specifically includes: The first model is retrained based on the specified training data to obtain the retrained second model. Construct test data based on the specified training data; The second model is tested using the test data to obtain the model evaluation index corresponding to the second model; Determine whether the model evaluation index is greater than a preset evaluation index threshold; If so, the second model shall be used as the test prediction model.

6. The data verification method based on artificial intelligence according to claim 1, characterized in that, After the step of verifying the specified landing result based on the prediction result and generating a verification result corresponding to the specified landing result, the method further includes: A data analysis report corresponding to the maintenance features is generated based on the verification results; Obtain the communication information of preset testers; Based on the communication information, the data analysis report is sent to the testers.

7. A data verification device based on artificial intelligence, characterized in that, include: The first judgment module is used to determine whether a data verification request triggered by the user has been received. The first acquisition module is used to acquire the policy number input by the user and the target dimension scene features selected by the user if the condition is met. The second acquisition module is used to acquire maintenance characteristics corresponding to the target dimension scenario features based on the policy number; The third acquisition module is used to acquire the specified database entry result corresponding to the maintenance feature based on the policy number; The prediction module is used to perform prediction processing on the target dimension scene features and the maintenance characteristics through a preset test prediction model, and generate prediction results corresponding to the maintenance characteristics. The first generation module is used to perform verification processing on the specified landing result based on the prediction result, and generate a verification result corresponding to the specified landing result. The third acquisition module includes: The second calling submodule is used to call the policy table corresponding to the policy number; The second filtering submodule is used to perform data query processing on the policy table based on the maintenance characteristics, and to filter out the specified characteristics that match the maintenance characteristics from the policy table; The second acquisition submodule is used to acquire the characteristic result value corresponding to the specified characteristic from the policy table; The second determining submodule is used to use the characteristic result value as the specified database entry result corresponding to the maintenance characteristic; In this system, a mapping relationship between different features and a policy table is pre-established, and the policy table is used to obtain the result value of the features; the specified feature that matches the maintenance feature mentioned above specifically refers to the feature stored in the policy table that is the same as the maintenance feature.

8. A computer device comprising a memory and a processor, wherein the memory stores computer-readable instructions, and the processor, when executing the computer-readable instructions, implements the steps of the data verification method based on artificial intelligence as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the data verification method based on artificial intelligence as described in any one of claims 1 to 6.

Images