An Improved Genetic Algorithm for Test Cases Generation Oriented Paths  

作　　者：MEI Jia WANG Shengyuan 

机构地区：[1]Department of Computer Science and Technology,Tsinghua University

出　　处：《Chinese Journal of Electronics》2014年第3期494-498,共5页电子学报（英文版）

基　　金：supported in part by National Natural Science Foundation of China(No.61272086,No.61170051);Guangxi Nature Science Fund(No.2012jj BAG0074);Guangxi Education Department Scientific Research Items(No.201106LX840);Guangxi Key laboratory of hybrid computation and IC design analysis Open Foundation(No.2012HCIC05);National Social Science Fund(No.11CTQ008)

摘　　要：This paper discusses a method that can automatically generate test cases for selected paths using a special genetic algorithm. The special algorithm is called Queen-bee evolutionary genetic algorithm(QBEA). In this algorithm, sequences of operators iteratively executes for test cases to evolve to target paths. The best chromosome called queen among the current population is crossover with drones selected according to a certain crossover probability, which enhances the exploitation of searching global optimum. A comparative experiment results prove that the proposed method is actually a great improvement in optimization efficiency and optimization effect.This paper discusses a method that can automatically generate test cases for selected paths using a special genetic algorithm. The special algorithm is called Queen-bee evolutionary genetic algorithm（QBEA）. In this algorithm, sequences of operators iteratively executes for test cases to evolve to target paths. The best chromosome called queen among the current population is crossover with drones selected according to a certain crossover prob- ability, which enhances the exploitation of searching global optimum. A comparative experiment results prove that the proposed method is actually a great improvement in opti- mization efficiency and optimization effect.

关 键 词：Software testing Test data generation Queen-bee evolutionary genetic algorithm. 

分 类 号：TP311.53[自动化与计算机技术—计算机软件与理论] TP18[自动化与计算机技术—计算机科学与技术]