Inferential analysis of Amazon’s top 50 best selling books

People today lack the time to go through the synopsis and prologue of every book to determine if they genuinely wanted to read it or not. This may cause missing reading of some really good books for a reader. Hence, books’ rating (based on choice of other readers) is made available to the readers so that they can decide which books to read necessarily. This review or rating of the book is generated based on the genre and author of the book. In this study, authors have implemented machine learning models like linear regression and logistic regression for the same. Metrics such as precision-recall curve and AUC-ROC curve are used to determine the rating of the book using different datasets from the data frame. The experimental evaluation is done in the Google Collaboratory platform where authors aim to evaluate the books’ reviews using a numerical dataset. The results obtained during the experimental evaluation are encouraging and hence advocate the implementation of such models at large.