Clustering Based Recommender System for Compilation of Research Papers

Due to addition of the journals and conference proceedings around the globe in the literature, it is observed that lots of research papers are daily published in the various categories of the journals and conference proceedings. Researchers sometimes encounter difficulties while endeavoring to comprehensively analyze all research publications for a particular field of study to locate pertinent studies which may be used for future scope of the work over the research topic. Hence, in the present work, a recommender system is explained to the research scholars by proposing articles based on assessments supplied by other academics within the similar domain of research. Collaborative filtering technique is used for the development of the recommender system, and it may be extensively utilized in several commercial recommender systems. It is obvious that the computational complexity of methods grows and directly proportion to the number of users and items. To address the said issues related to scalability, a proficient recommender system is presented that makes use of subspace clustering. The approach entails examining the researcher-paper matrix to ascertain the correlations among different researchers. Through the relationships among the keywords of the research papers, the present work offers a well selected compilation of research articles for recommendation which may be used for future research work.