Books are said to be the best friend a person can have. Book reading culture dates back to almost a couple of thousands of years. After ancient civilizations learned to write, they stored information in tablets or walls or stones are said to be the predecessors of books. The newest form of books is called e-books, digitalization or digital printing of paper-based books. In Bangladesh, even a few years back, people had to go to the library in person to collect books but now online book stores are getting popular. With all its perks being easy, online book store comes with some penalty i.e. reader don't know about the books or the service of the book store itself. To avoid such, book readers tend to rely on reviews and ratings. Our goal is to analyze Bangla language reviews and give proper feedback about books and the online store so the book readers can buy desired books to read and get better services from online book stores. 6281 raw data are collected to train the machine. In this study, the Machine Learning and Deep Learning methods are used for the binary (negative, positive) analysis of sentiments. Natural Language Processing (NLP) is used for text preprocessing. Some of the popular algorithms like Support Vector Machine (SVM), Decision Tree, KNN, Random Forest, Logistic Regression and Long Short-Term Memory(LSTM) are used to classify the sentiments. Our Highest accuracy is 97.49%, achieved by LSTM.