Personalized Book Recommendation System using Machine Learning Algorithm

As the amounts of online books are exponentially increasing due to COVID-19 pandemic, finding relevant from a vast e-book space becomes tremendous challenge for users. Personal recommendation systems have been emerged conduct effective search which mine related based on user rating and interest. Most these existing user-based ratings where content-based collaborative-based learning methods used. These systems' irrationality is their technique, counts users who already unsubscribed services no longer rate books. This paper proposed an system recommending that rated book using clustering method then found similarity suggest new book. The used K-means Cosine Distance function measure distance Similarity find between clusters. Sensitivity, Specificity, F Score were calculated ten different datasets. average Specificity was higher than sensitivity, means classifier could re-move boring reader's list. Besides, receiver operating characteristic curve plotted graphical view classifiers' accuracy. datasets close ideal diagonal line far worst line. result concludes recommendations, particular book, more accurately system.