An early software-quality classification based on improved grey relational classifier

摘要: The inherent uncertainty and incomplete information of the software development process presents particular challenges for identifying fault-prone modules providing a preferred model early enough in cycle order to guide enhancement efforts effectively. Grey relational analysis (GRA) grey system theory is well known approach that utilized generalizing estimates under small sample uncertain conditions. This paper examines potential benefits an software-quality classification based on improved classifier. particle swarm optimization (PSO) adopted explore best fit weights metrics GRA deriving classifier with balance misclassification rates. We have demonstrated our by using data from medical dataset. Empirical results show proposed provides rates than classifiers without PSO. It also outperforms widely used regression trees (CART) C4.5 approaches.