Multi Cloud Based Service Recommendation System Using DBSCAN Algorithm

Abstract

The historical collaborative filtering based recommendation system has become the most essential one in cloud computing environment. The recommender system which has been depends on collaborative filtering easily identify the user references and allow to learn relationship items—past users from the user group who exhibit similar preferences. Hence, the recommendation system has been considered as the most powerful tool for cloud providers and users. This paper proposed the clustering recommendation system executed in cloud environment. The accuracy of the system reduced when irrelevant features presents in data. So that in this proposed scheme, an effective feature selection approach named as modified LDA has been utilized for acquiring the relevant information only. The LDA technique defined as Linear Discriminant Analysis which decreased features numbers to expected value before classification process. DBSCAN is utilized as a clustering approach which provides better quality in terms of segregating the number of movies. Based on the genre, the similar movies are clustered together with the user ratings. DBSCAN elaborated as Density Based Spatial Clustering of Applications with Noise termed as famous method of learning or clustering which detached the high density clusters from low density clusters. The LDA technique has been utilized to accomplish the appropriate user reviews with categorized ratings. In this paper, the main challenge described as to analyze the reviews of the user with ratings using best clustering approach. The evaluation results proved the highest accuracy of 92.56% when compared with various methods of accuracy. This proposed method also has minimum execution time.