Role of Data Mining and Machine Learning in Software Reusability

Machine learning and data mining have opened new streams in research and software development. Likewise, integration of machine learning with software development life cycle has provided opportunities for well-organized and effective development. Software reusability is an important aspect of SDLC. Therefore, automation of software reusability plays a dynamic role in SDLC. It curtails the cost and effort required for development of a software product. This paper provides a taxonomical mapping of reuse metrics and corresponding machine learning techniques applicable. Artificial intelligence techniques i.e., neural networks and clustering are nominated in the paper to check their applicability for automation of software reusability. A model is created to identify best machine learning practice amongst the nominated techniques for automation of software reusability. Evaluation of model based on survey results identifies ANN: the best technique among neural networks and clustering.