Abstract: Owing to the exponential growth of information in online social networks, the users of such networks demand the recommendation systems to deliver significant results. A recommendation system rightly suggests the personalized movies that are desirable to the users predominantly from large information storage. Notably, the current research works in movie recommendation system focus on determining the most relevant features from the user profile information and shared contents in the social network. Even though the existing research works recommend the movies that are in proximity to the user preferences, there is a profound need for further exploring the features of the movie and thus ensure the highly desired movies to the users. Hence, this paper targets on recommending the movies with the knowledge of analyzing the movie features along with the data clustering and computational intelligence methods. This article proposes the Cuckoo search based MOst personalized VIEw in item recommendation (CMOVIE) model, incorporating the missing rating prediction and contextual movie recommendation phases. At first, the C-MOVIE approach explores the features of the movies to recognize the interest of the users in terms of inherent features after reducing the feature dimensionality by Principal Component Analysis (PCA) method. Then, it clusters the users based on the recognized features by K-means clustering and Cuckoo search optimization methods with the intention of grouping the users with similar interests which eases the missing rating prediction when using Probabilistic Matrix Factorization (PMF). In the end, the C-MOVIE approach contextually recommends the movies to the users by mapping the features of the new movies with the features of the clustered users. The experimental results yielded on Douban movie which data set demonstrate that the CMOVIE approach distinctively delivers the personalized movie recommendation than the existing HPSO method.
S. Uma Shankari and M. Chidambaram, 2020. Cuckoo Search Based Personalized View for Movie Recommendation over Social Networks. Research Journal of Applied Sciences, 15: 102-111.