Using Machine Learning for Improved Book Recommendations Based on User Insights

Abstract

In this work recommendation system, the Goodread dataset is analysed using advanced data analysis techniques. Initially, the preprocessing of the data is performed to ensure quality and consistency in the text. Next, the insights into highly rated books, patterns of user engagement, and language distribution are evaluated to get more information about the distribution. Finally, the K-Means clustering of the dataset with the optimized number of clusters is evaluated showing meaningful grouping and highlighting how the book dataset forms various clusters depending on the rating and popularity. A weighted rating system is used to remove biases from the ranking of books. This study gives a practical implication for publishers, authors, and readers to understand market trends and make personalized recommendations.