Book Recommendation for eLearning Using Collaborative Filtering and Sequential Pattern Mining

Reading is a fundamental skill that every person needs to expand throughout his/her lifetime. Book recommendation for eLearning systems can gain more attention in digital libraries, commercial websites and social media sites. Nowadays, obtaining the preferred book in real-time becomes a challenging task. Because there are too many books available online and offline, this colossal number creates a dilemma for the end-user, especially for ebook learners. The book recommendation helps overcome the information overhead problem and through this, the user can quickly get books according to need within the shortest span of time. This paper proposes a Book recommendation system for eLearning using collaborative filtering and sequential pattern mining to serve individual requirements. Current writing also focuses on various similarity techniques, namely Cosine, Euclidean, Correlation, Manhattan and Jaccard. The accuracy of the Book Recommendation for eLearning is evaluated by applying precision, recall and F1 Score. In the end, results show that correlation-based Sequential pattern mining gives a better F1 Score.