The Optimized Selection Research Of Regression Test Cases

Regression testing is a frequent and expensive process in the software life cycle,study the optimal selection of test cases is to ensure the good coverage premise andhigher test efficiency. At present, the optimized selection research of regression testcases hot spot mainly concentrated in the application of some optimization algorithmand interaction strategies in the test cases focus on search in accordance withestablished standards, minimizing the cover of the example set by a test, the structuraloptimization model, in order to eliminate redundant test cases, cut to the execution ofthe test suite scale, thus reduce test cost, manpower and material resourcesconsumption. In recent years, although the software regression test cases optimizationtechnology and method has great development, it is still far from mature, theoptimization of the existing technology still can’t satisfy the current software testingpractical needs. In order to ensure the quality and full test software, the premise ofhow to use less as far as possible the test cases to complete a regression test, so as toreduce the cost of software testing, to improve test efficiency is a very valuableresearch subject.Traditional heuristic algorithm for its theoretical basis is simple, the operation issimple, robust, in a regression test cases optimization areas to be a certain degree ofapplication and promotion. But in the past the literature study, scholars often think theNP-HARD problem is difficult to use traditional algorithm to solve, even if thesolution is to get the local optimal solution, rather than the global optimal solution.This article in this conclusion based on the question, trying to explore the traditionalalgorithm based on improved optimization algorithm to solve the optimizationproblem breakthrough the technical bottleneck.This thesis focus on a regression test, first at home and abroad were reviewed insoftware regression test cases optimization choose technical work, and based on theexisting some result, software test cases to return the optimized selection oftechnology is studied, the contrast at present various optimized algorithm, especiallythe five classical traditional heuristic algorithm and the characteristics of theapplicable domain, analysis the advantages and disadvantages of different algorithmand the optimization principle. Key summarizes the these traditional algorithm failedto get the global optimal solution of the reason. In the existing research results based on the proposed based on a traditional H ofthe return of the improved algorithm test cases of the optimal selection of newalgorithm, this algorithm in clear H algorithm based on the details about Jane, addalternate find necessary test cases, and delete1-1redundant test cases and delete theleading position of the low test demand of the steps. And give the specific algorithmis false code. Finally through the algorithm analysis and experimental results showthat from generated test suite size and test generation time consumption of twoevaluation, the new method is better than the existing related algorithm.