ABSTRACT
The internet is pervasive and people express and share their personal thoughts in the web environment. The rising number of web 3.0 tools in recent years has accumulated huge amounts of data dynamically. Analyzing these data and understanding the behavioral pattern helps in dynamic changes in all fields. The technique of sentiment analysis of data has gained significant attention in recent years as it can help in obtaining valuable insights from online reviews. With the advancement in technology, e-books and digital reading platforms like Kindle has gained popularity. Analyzing the sentiments from Kindle book reviews has become an important area of research as it can help in giving feedback to authors and for product improvement. This chapter presents a substantive evidence-based study on analyzing sentiments of Kindle book reviews using deep learning techniques. The study employs a comprehensive dataset consisting of a diverse range of Kindle book reviews collected from various genres and authors. The dataset is preprocessed to remove noise, including irrelevant information and duplicate entries, ensuring the validity and reliability of the results. The sentiment analysis task involves classifying reviews into positive, negative, or neutral categories, providing an overall sentiment rating for each review. The findings highlight the potential of LSTM and bidirectional LSTM in extracting meaningful sentiment information, enabling a deeper understanding of readers’ reactions and preferences
