A Personalized Time-Sequence-Based Book Recommendation Algorithm for Digital Libraries

Book recommendations are of great significance in colleges and universities. Although current recommendation approaches have made significant achievements, these approaches do not consider college students' similar learning trajectories in the same major. In order to recommend books more accurately, mining the knowledge system is very crucial for college students in the same major. This paper proposes a personalized book recommendation algorithm that is based on the time sequential collaborative filtering recommendation, combined with students' learning trajectories. In order to recommend books effectively, our algorithm leverages space distance. In this algorithm, we consider two important characteristics: the time sequence information of borrowing books and the circulation times of books. Our experimental results demonstrate that our book recommendation algorithm is in accordance with the college students' demand for professional learning.