ABSTRACT
An important attribute in software quality assurance system is software reliability. Since several years, a number of works have been done to improve software reliability. The major tasks involved in enhancing software reliability include software improvement, software modelling and software measurement. For evaluating the reliability of software, a number of techniques have been proposed in this chapter, but machine learning proves to be an excellent technique for generating reliability by parameter evaluation. Several ML techniques have been evolved for capturing several software systems. In this chapter, we focus on building an item-item recommender system using collaborative filtering. Our proposed model uses the well-known MovieLens data set and also uses the concept of Bayesian average for evaluating movie popularity. In order to deal with the problem of sparsity, our proposed model builds compressed sparse row (CSR) matrix. Finally, our model uses machine learning approach using k-nearest neighbours for recommending movies based on similarity.
