Eliminating Malicious Feedback given by Cloud Customer Using Machine Learning Techniques to Enhance Trust

Abstract

Machine Learning techniques have been used significantly for identifying malicious feedbacks. This work has been done to compare machine learning techniques on the feedback given by the customer to the cloud service provider. Analysis using machine learning techniques was done on the dataset of Cloudarmor using four machine learning techniques, namely SVM, Random Forest Classifier, Naïve Bayes and Decision Tree Classifier. Analysis of the feedback has been done on Trust Result calculated on the parameters of Availability, Price and Technical Support. Out of the above mentioned techniques, Random Forest Classifier depicted best outcome, showing values for SVM, Random Forest Classifier, Naïve Bayes and Decision Tree Classifier as 95.27, 96.92, 90.54 and 96.80 respectively.