Trust Evaluation and Cloud Service Recommendation System Based On Machine Learning Techniques

Abstract

Cloud computing is a vast platform which provides users with ample services. It offers various benefits such as expense reduction, resource elasticity, managing data and easier services for hosting. At the same time, the aspects such as security, performance and reliability are the key concerns in cloud paradigm. Therefore, trustworthiness in adopting a cloud service becomes an important issue from the end user's viewpoint. Trust represents the degree to which user's expectations about the capabilities of a service are met. This paper proposes trust evaluation and cloud service recommendation system (TECSRS) which concentrates on evaluation of trust using Quality of Service (QoS) parameters. TECSRS uses Machine Learning techniques which include Linear Regression, Artificial Neural Networks (ANN), Random Forest Regression and Support Vector Regression (SVR), in trust computation. TECSRS facilitates recommending a cloud service to the end user and forecasting the trust values of cloud services. Experimental results depict that the Random Forest Regression model outperforms the other machine learning models with regard to accuracy of trust computation.