CN-121998773-A - Data processing method, device, computer equipment and medium based on artificial intelligence

CN121998773ACN 121998773 ACN121998773 ACN 121998773ACN-121998773-A

Abstract

The application belongs to the technical field of artificial intelligence, and relates to a data processing method, a device, computer equipment and a storage medium based on artificial intelligence, wherein the method comprises the steps of reasoning business data based on a business model to obtain a prediction result; the method comprises the steps of obtaining a prediction result by using an anomaly discovery agent based on context information corresponding to service data, obtaining an anomaly detection result by using the anomaly discovery agent, obtaining an inspection result by reasonably detecting the service data based on the data inspection agent if the anomaly detection result is abnormal, obtaining a target prediction result by correcting the service data based on the inspection result by using a correction agent, and outputting the target prediction result if the target prediction result passes verification processing based on a verification strategy. The application can be applied to the result analysis scene related to the business model in the field of financial science and technology, improves the processing efficiency of misjudgment and identification of the predicted result of the business model, and ensures the accuracy of the generated correction result.

Inventors

WANG JIANZONG
ZHANG NAN
QU XIAOYANG

Assignees

平安科技（深圳）有限公司

Dates

Publication Date: 20260508
Application Date: 20260109

Claims (10)

1. A data processing method based on artificial intelligence, comprising the steps of: acquiring business data to be processed, and carrying out reasoning processing on the business data based on a preset business model to obtain a corresponding prediction result; acquiring context information corresponding to the service data; Based on the context information, performing abnormality detection processing on the prediction result by using a preset abnormality discovery agent to obtain a corresponding abnormality detection result; If the abnormality detection result is that abnormality exists, carrying out rationality detection processing on the service data based on a preset data examination intelligent agent to obtain a corresponding examination result; Based on the examination result, carrying out result correction processing on the service data by using a preset correction agent to obtain a corresponding target prediction result; Performing verification processing on the target prediction result based on a preset verification strategy; and if the target prediction result passes the verification, outputting the target prediction result.
2. The method for processing data based on artificial intelligence according to claim 1, wherein the step of performing anomaly detection processing on the prediction result by using a preset anomaly discovery agent based on the context information to obtain a corresponding anomaly detection result specifically comprises: calculating confidence distribution entropy corresponding to the prediction result based on the abnormal discovery agent; Calculating a confidence margin index corresponding to the prediction result; performing anomaly detection calculation processing on the business data and the context information based on a preset anomaly detection model to obtain a corresponding abnormal value; Calculating the confidence distribution entropy, the confidence margin index and the abnormal value based on a preset abnormal score calculation strategy to obtain a corresponding abnormal score; Judging whether the abnormality score is larger than a preset abnormality threshold; If so, generating a first abnormality detection result with an abnormality corresponding to the prediction result, otherwise, generating a second abnormality detection result without an abnormality corresponding to the prediction result.
3. The method for processing data based on artificial intelligence according to claim 1, wherein the step of obtaining the corresponding inspection result by performing the rationality detection processing on the service data by the data inspection agent based on the preset data inspection agent specifically comprises: Based on the data examination intelligent agent, carrying out knowledge retrieval on the business data by using a preset knowledge base to obtain a corresponding knowledge retrieval result; extracting fact data from the knowledge retrieval result; carrying out semantic comparison on the prediction result and the fact data based on a preset first large language model to obtain a corresponding semantic comparison result; Carrying out causal consistency test on the predicted result to obtain a corresponding test result; report generation processing is carried out based on the knowledge retrieval result, the semantic comparison result and the inspection result, so as to obtain a corresponding target report; reporting the target as the examination result.
4. The method for processing data based on artificial intelligence according to claim 1, wherein the step of performing result correction processing on the service data by using a preset correction agent based on the examination result to obtain a corresponding target prediction result specifically comprises: Inquiring historical similar cases corresponding to the business data from a preset case library based on the correction agent; Analyzing and processing the historical similar cases and the examination results based on a preset second large language model to generate a plurality of corresponding candidate results; Performing similarity calculation processing on each candidate result based on a preset multi-source consistency verification strategy to obtain a plurality of corresponding similarity scores; screening out target candidate results with highest similarity scores from all candidate results; and taking the target candidate result as the target prediction result.
5. The method for processing artificial intelligence-based data according to claim 4, wherein the step of performing similarity calculation processing on each candidate result based on a preset multi-source consistency verification policy to obtain a plurality of corresponding similarity scores comprises: calculating a first similarity between a specified candidate result and a preset knowledge base rule, wherein the specified candidate result is any one result of all the candidate results; Calculating a second similarity between the specified candidate result and an interpretation vector of the business model; calculating a third similarity between the specified candidate result and the label result of the historical similar case; Weighting calculation processing is carried out on the first similarity, the second similarity and the third similarity based on a preset comprehensive calculation formula, and corresponding calculation results are obtained; And taking the calculated result as a specified similarity score of the specified candidate result.
6. The method for processing data based on artificial intelligence according to claim 1, wherein the step of performing verification processing on the target prediction result based on a preset verification policy specifically comprises: Calling a preset verification model; Performing verification loss calculation processing on the service data and the target prediction result based on the verification model to obtain a corresponding loss value; judging whether the loss value is smaller than a preset threshold value or not; If yes, judging that the target prediction result passes the verification, otherwise, judging that the target prediction result does not pass the verification.
7. The artificial intelligence based data processing method according to claim 1, further comprising, after the step of performing output processing on the target prediction result: Calling a preset data flywheel executing module; screening out target sample data meeting preset reflux conditions based on the data flywheel executing module; performing light fine tuning training on the service model based on the target sample data to obtain a corresponding target service model; Performing contrast test on the target service model based on a preset test strategy; And if the target service model passes the comparison test, carrying out model updating processing on the service model based on the target service model.
8. An artificial intelligence based data processing apparatus comprising: The reasoning module is used for acquiring the business data to be processed, and carrying out reasoning processing on the business data based on a preset business model to obtain a corresponding prediction result; the acquisition module is used for acquiring the context information corresponding to the service data; The first detection module is used for carrying out abnormality detection processing on the prediction result by using a preset abnormality discovery intelligent agent based on the context information to obtain a corresponding abnormality detection result; the second detection module is used for carrying out rationality detection processing on the service data based on a preset data examination agent if the abnormality detection result is that an abnormality exists, so as to obtain a corresponding examination result; the correction module is used for carrying out result correction processing on the business data by using a preset correction agent based on the examination result to obtain a corresponding target prediction result; The verification module is used for verifying the target prediction result based on a preset verification strategy; and the output module is used for outputting the target prediction result if the target prediction result passes the verification.
9. A computer device comprising a memory having stored therein computer readable instructions which when executed implement the steps of the artificial intelligence based data processing method of any of claims 1 to 7.
10. A computer readable storage medium, characterized in that it has stored thereon computer readable instructions which, when executed by a processor, implement the steps of the artificial intelligence based data processing method according to any of claims 1 to 7.

Description

Data processing method, device, computer equipment and medium based on artificial intelligence Technical Field The application relates to the technical field of artificial intelligence, and can be applied to the field of financial science and technology, in particular to a data processing method, a data processing device, computer equipment and a storage medium based on artificial intelligence. Background In the intelligent insurance and claim settlement auditing scene in the financial insurance field, although the existing business model can generate a prediction result based on input data, the accuracy verification and correction mechanism still has obvious defects. In particular, the prior art generally only relies on manual case-by-case review of model output, which is low in efficiency and easily influenced by subjective experience, accuracy cannot be guaranteed, large-scale application is difficult to realize, and meanwhile, due to the lack of dynamic identification and self-correction capability of prediction errors, the misjudgment rate of the model in complex service scenes (such as non-standard body verification and high fraud risk claim settlement) is high, so that an insurance company needs to bear additional claim cost or customer experience is damaged. For example, in the health insurance claim audit, the existing model may misjudge the symptomatically related treatment cost of the present application claim rejected as a payable item because of insufficient association between the historic visit record of the insured person and the current claim application, and if the association is not found to be missed during the manual review, the fund loss of the insurer is directly caused. Therefore, it is needed to develop an intelligent auditing system with real-time misjudgment, identification and self-correction capabilities so as to improve the accuracy and automation level of financial insurance business decisions. Disclosure of Invention The embodiment of the application aims to provide a data processing method, a device, computer equipment and a storage medium based on artificial intelligence, which are used for solving the technical problems that the efficiency is low and the accuracy cannot be ensured in the existing mode of manually checking model output case by case. In a first aspect, there is provided an artificial intelligence based data processing method, comprising: acquiring business data to be processed, and carrying out reasoning processing on the business data based on a preset business model to obtain a corresponding prediction result; acquiring context information corresponding to the service data; Based on the context information, performing abnormality detection processing on the prediction result by using a preset abnormality discovery agent to obtain a corresponding abnormality detection result; If the abnormality detection result is that abnormality exists, carrying out rationality detection processing on the service data based on a preset data examination intelligent agent to obtain a corresponding examination result; Based on the examination result, carrying out result correction processing on the service data by using a preset correction agent to obtain a corresponding target prediction result; Performing verification processing on the target prediction result based on a preset verification strategy; and if the target prediction result passes the verification, outputting the target prediction result. In a second aspect, there is provided an artificial intelligence based data processing apparatus comprising: The reasoning module is used for acquiring the business data to be processed, and carrying out reasoning processing on the business data based on a preset business model to obtain a corresponding prediction result; the acquisition module is used for acquiring the context information corresponding to the service data; The first detection module is used for carrying out abnormality detection processing on the prediction result by using a preset abnormality discovery intelligent agent based on the context information to obtain a corresponding abnormality detection result; the second detection module is used for carrying out rationality detection processing on the service data based on a preset data examination agent if the abnormality detection result is that an abnormality exists, so as to obtain a corresponding examination result; the correction module is used for carrying out result correction processing on the business data by using a preset correction agent based on the examination result to obtain a corresponding target prediction result; The verification module is used for verifying the target prediction result based on a preset verification strategy; and the output module is used for outputting the target prediction result if the target prediction result passes the verification. In a third aspect, a computer device is provided, comprising a memory, a processor and