Pre-Recommendation Clustering and Review Based Approach for Collaborative Filtering Based Movie Recommendation

The recommendation is playing an essential\udpart in our lives. Precise recommendations facilitate users\udto swiftly locate desirable items without being inundated\udby irrelevant information. In the last few years, the\udamount of customers, products and online information\udhas raised speedily and results out into the huge data\udanalysis problem for recommender systems. While\udhandling and evaluating such large-scale data, usual\udservice recommender systems regularly undergo\udscalability and inefficiency problems. Nowadays, in\udmultimedia platform such as movie, music, games, the\uduse of Recommender System is increased. Collaborative\udFiltering is a dominant filtering technique used by many\udRSs. CF utilizes the rating history of the user to find out\ud“like minded” users and this set of like-minded user is\udthen used to recommend the movies which are liked by\udthese like-minded users but did not watch by the active\uduser. Thus, in CF, to find out the “neighborhood” the\udrating history of a user is used, but the reason behind the\udrating is not considered at all. This will lead to inaccuracy\udin finding a neighborhood set and subsequently in\udrecommendation also. To cope with these scalability and\udaccuracy challenges, this paper proposes an innovative\udsolution, Clustering and Review based Approach for\udCollaborative Filtering based Recommendation. This\udinnovative approach is enacted with the two stages; in the\udfirst stage the clustering of the available movies for\udrecommendation is clustered into the subclasses for\udfurther computation. In the succeeding stage, the\udmethodology based on reviews is utilized for finding\udneighborhood set in User Based Collaborative Filtering