A further analysis on the use of Genetic Algorithm to configure Support Vector Machines for inter-release fault prediction

Abstract

Some studies have reported promising results on the use of Support Vector Machines (SVMs) for predicting fault-prone software components. Nevertheless, the performance of the method heavily depends on the setting of some parameters. To address this issue, we investigated the use of a Genetic Algorithm (GA) to search for a suitable configuration of SVMs to be used for inter-release fault prediction. In particular, we report on an assessment of the method on five software systems. As benchmarks we exploited SVMs with random and Grid-search configuration strategies and several other machine learning techniques. The results show that the combined use of GA and SVMs is effective for inter-release fault prediction.