Model-Based Book Recommender Systems using Naive Bayes enhanced with Optimal Feature Selection

Book recommender systems play an important role in book search engines, digital library or book shopping sites. In the field of recommender systems, processing data, selecting suitable data features, and classification methods are always challenging to decide the performance of a recommender system. This paper presents some solutions of data process, feature and classifier selection in order to build an efficient book recommender system. The Book-Crossing dataset, which has been studied in many book recommender systems, is taken into account as a case study. The attributes of books are analyzed and processed to increase the classification accuracy. Some well-known classification algorithms, such as, Naive Bayes, decision tree, etc., are utilized to predict user interests in books and evaluated in several experiments. It has been found that Naive Bayes is the best selection for book recommendation with acceptable run-time and accuracy.