CSOFS: Feature Selection Using Cuckoo Search Optimization Algorithm for Software Fault Detection

The method of identifying and predicting software flaws or problems in software systems involves using historical data. Software fault prediction is a crucial topic of research in the field of software engineering because it enables developers to find probable flaws early on, which can save money and enhance the quality of the programme produced. The study’s objective is to determine how well the Cuckoo Search Optimization Algorithm can extract the features that are most important for correctly predicting software flaws. The prediction models were tested and trained using a dataset of software metrics, and the dataset was subjected to a variety of feature selection algorithms. We have performed an analysis by comparing the efficiency of various optimization algorithms namely evolutionary based Genetic Algorithm Feature Selection (GAFS), Differential Evolution Feature Selection (DEFS), and swarm based algorithms like Ant Colony Optimization Feature Selection (ACOFS) and Particle Swarm Optimization Feature Selection (PSOFS) with respect to Cuckoo Search Optimization Algorithm (CSOFS). The results from the output tables for various metrics prove that feature selection using CSO can significantly increase the accuracy of the predictive model, as well as reduce the complexity of the models. This study’s findings have crucial implications for developing a software failure prediction models that is more accurate and efficient.