Book Recommendation System using Machine learning

EasyChair Preprint no. 4330

Abstract

Machine learning is a scientific study of statistical model and algorithms. In this research I will use the machine learning algorithms, K-NN and matrix factorization. In the books recommendations system BX books dataset is used. Suggestion method is a selection strategy which was used for collective selection and material-based sorting strategies. Pattern filtering technique is carried to suggest a consumer to an element the "rank" or "first option." Suggestion process collected information was about either the customer's first option on unusual subject relevant to films, books, travel, TV and commerce, etc. And from the other side, an effective selection of books recommendation system design utilizes prior scores or background of the customer. Cooperative sorting is a process of measuring and processing the categories across user opinions. Cooperative filtering first gathers the rankings or a preference of books provided by multiple users and then suggests books to different individuals based on various previous tastes and preferences. K-Means Multipathing together with K-Nearest Neighbor is applied on the BX dataset to achieve the greatest-optimized outcome. In prior methodology, the information is dispersed and ends in a high amount of matrices, whereas the information is collected throughout the suggested strategy and concludes in a small number of groupings. The preferred framework forecasts the customer's desire for a book based on various criteria. These consumers will affect their views on one another. It maximizes the succession and has smaller RMSE.

Keyphrases: book recommendation, collaborative filtering, KNN, matrix factorizations., Recommender System

@Booklet{EasyChair:4330,
  author = {Fatima Ijaz},
  title = {Book Recommendation System using Machine learning},
  howpublished = {EasyChair Preprint no. 4330},

  year = {EasyChair, 2020}}