Software analysis using cuckoo search

Abstract

Software analysis includes both Code coverage as well as the Requirements coverage. In code coverage, automatic test sequences are generated from the control flow graph in order to cover all nodes. Over the years, major problem in software testing has been the automation of testing process in order to decrease overall cost of testing. This paper presents a technique for complete software analysis using a metaheuristic optimization technique Cuckoo Search. For this purpose, the concept of Cuckoo search is adopted where search follows quasi-random manner. In requirement coverage, test sequences are generated based on state transition diagram. The optimal solution obtained from the Cuckoo Search shows that it is far efficient than other metaheuristic techniques like Genetic Algorithm and Particle Swarm optimization.