基于智能算法的软件测试方法研究  

Research on Software Testing Methods Based on Intelligent Algorithms

作　　者：杨春静 杨达 葛长赟 YANG Chunjing;YANG Da;GE Changyun(Dalian Neusoft University of Information,Dalian Liaoning 116023)

机构地区：[1]大连东软信息学院,辽宁大连116023

出　　处：《软件》2023年第5期32-34,共3页Software

基　　金：2022年辽宁省教育厅基本科研项目青年项目(LJKQZ20222441)。

摘　　要：随着计算机技术的不断发展,软件规模和复杂度增加导致软件质量问题越来越严重。传统的软件测试方法存在测试覆盖率低、测试准确性差等问题。智能算法被提出作为一种新型的软件测试方法,可以提高测试效率和准确性。遗传算法可以自动化生成测试用例,具有高覆盖率和测试效率;神经网络则利用学习能力提高测试准确性和效率。智能算法还可以应用于缺陷检测和测试用例优化。未来,智能算法在软件测试中的应用还有很大的发展空间。With the continuous development of computer technology,the increase in software scale and complexity has led to increasingly severe software quality problems.Traditional software testing methods have problems such as low test coverage and poor accuracy.Intelligent algorithms have been proposed as a new type of software testing method,which can improve testing efficiency and accuracy.Genetic algorithms can automate the generation of test cases,with high coverage and efficiency,while neural networks use learning ability to improve testing accuracy and efficiency.Intelligent algorithms can also be applied to defect detection and test case optimization.In the future,the application of intelligent algorithms in software testing still has great development space,which can reduce errors and costs in software development.

关 键 词：智能算法 软件测试 遗传算法 神经网络 测试用例优化 缺陷检测 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]