Intelligent test case generation method based on AI confrontation model

By alternating training of the AI ​​adversarial model generator and discriminator, the problem of insufficient intelligence in test case generation and analysis in existing technologies is solved. The entire process of test case generation, execution and analysis is automated and intelligent, which improves testing efficiency and quality and discovers hidden defects that are difficult to find by traditional methods.

CN122261990APending Publication Date: 2026-06-23BEIJING ZHENGYUAN ZHISHENG TECHNOLOGY CO LTD
3 Cites 0 Cited by

Patent Information

Application Number
CN202610214443.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-02-14
Publication Date
2026-06-23

AI Technical Summary

Technical Problem

In existing technologies, test case generation often relies on rule bases or simple machine learning models, making it difficult to generate high-value test cases that are adversarial and have boundaries. The intelligence level of test execution and result analysis is low, lacking adaptability, test coverage assessment is singular, and AI technology integration is insufficient, resulting in information silos.

Method used

By employing an AI adversarial model, high-quality test cases are generated through alternating training of the generator and discriminator. The model is then optimized through a dynamic feedback mechanism, achieving full automation and intelligence in the process of test case generation, execution, and analysis. Combined with multi-dimensional coverage evaluation, the comprehensiveness and effectiveness of test coverage are ensured.

Benefits of technology

The generated test cases are more adversarial and targeted, and can discover hidden defects that are difficult to detect by traditional methods. The testing process is adaptive and has learning capabilities, realizing the integration of AI technology throughout the entire process, significantly improving testing efficiency and quality, and reducing labor costs.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention relates to an intelligent test case generation method based on an AI confrontation model, which belongs to the technical field of computer software testing and comprises the following steps: constructing an initial training set according to tested system information, test requirements and historical test data; based on the initial training set, utilizing a generator of an AI confrontation model to generate a test case; evaluating the quality of the test case through a discriminator, and iteratively optimizing the generator based on an evaluation result; executing the generated test case, and recording an execution log and an intermediate result; and evaluating the execution log and the intermediate result until a preset test termination condition is met. The test case generated by using the AI confrontation model has higher confrontation and pertinence, and hidden defects which are difficult to perceive by a traditional method can be effectively found.
Need to check novelty before this filing date? Find Prior Art