Automatic Generation Of Test Cases Based On Improved Genetic Algorithm

The development of the automated testing is an inevitable trend for software testi Through the automated test tool to record script, playback and other operation methods can reduce the repetitive manual operations and the error rate greatly. However, the mos the automated test equipments are for test implementation, management and other w(?) but not for automated test case generation. During the software testing process, particul(?) in unit test testers still use manual methods to design and generate the required test ca(?) When the program is huge or has complex paths, writting test cases by hand will b heavy work, it is difficult to cover all the test paths and easy to be wrong, maybe result in low test efficiency and a great block on software development process.Automated test case generation is developed in this background and this met(?) improves above defects of hand-written test cases, in which the test cases automatic(?) generated by artificial intelligence algorithms has become the reserarch hot spots for high coverage and high efficiency in recent years. However, because the generation of (?) cases is an undecidable problem, and the program is usually huge and complex, the gen(?) search algorithm has been extremely limited. Genetic algorithm has a very clear advant(?) in dealing with uncertainly search problems. Genetic algorithm is a highly paral randomized, self-adaptive search algorithm which learnning from biological mechanis of natural selection and evolution. It has good global search capability, but for the lo search space is not very effective, easily leading premature convergence and falling i(?) local optimum.To resolve the above defects of genetic algorithm, this article improves automatically generated test cases system model on the basis of previous research firs(?) secondly, the article improves the select and crossover operator of genetic algorit(?) against the shortcomings of premature convergence, and combines the improved select(?) operator and the optimal preservation strategy for improving search capabilities in the lo space and the overall operating efficiency; Finally, the improved algorithm is applied generate test cases automatically and with specific examples, the article compares improved genetic algorithm and traditional genetic algorithm in terms of efficiency and effectiveness when generating the test cases. Experimental data shows that using the improved genetic algorithm to generating test cases automatically is feasible and efficient; Compared with the traditional genetic algorithm, the improved genetic algorithm has obvious advantages when generates test cases in both the efficiency and effectiveness.