Recommender System Based on Network Structure Link Analysis Technique Through Community Detection in Social Network to Handle the Cold-Start Problem

Abstract

The recommendation system plays a pivotal role in electronic commerce sites to trade their products and services to consumers. The system further helps in identifying the items for the user and might be worthful for creating buying options that are purely rely on recommender. Especially, in the case of a book recommender system, it is necessary to recommend the books to learn in conformity with the learning environment of the learner. The prime objective is to propose an efficient book recommender system for social networking communities. Moreover, the paper addresses all the important parameters required to compute the efficiency of the system. The authors also outlined existing approaches utilize in book recommendation system along with major findings. The author took one of the fundamental challenges of cold-start problem and proposed a new book recommender system’s framework using the “link analysis technique” of network structure. The proposed structure utilizes a hybrid technique for a book recommender system. Besides, the paper has tremendous potential to appraise the reader to explore a new book recommender system which further suggests the books to a novice on the social network as well as in-network communities.