Tuning the Library Performance

摘要

This paper describes a service for providing book recommendations, which is part of a digital library project whose principal goal is to develop technologies for supporting digital services. The proposed book recommendation system makes use of the usage logs of a digital library. The recommendation framework consists of three sequential steps: data preparation of the usage log, discovery of book associations using the pseudo rating matrix and book recommendations. The Pearson coefficient algorithm has been used here. In this paper, we present a novel method of building a collaborative filtering-based recommender system with higher accuracy for a library environment even in the absence of explicit feedback. The recommender systems for books provide personalized recommendations on books to users, who then spend less time searching for the right books. Our approach will be especially useful because we can build an effective recommender system based on collaborative filtering only using implicit feedback such as borrowed information and student detail.