ABSTRACT
Application distribution platforms such as Apple App Store and Google Play enable users to search, buy and install software applications with a few clicks. Additionally, application developers are working in different directions, which leads to the production of a variety of applications. This chapter provides techniques for classifying the reviews based on their metadata (star rating), keyword frequencies and sentiment analysis. It compares the accuracy of multiple Supervised machine learning (ML) classification algorithms. The chapter utilizes feature extraction (FE) algorithms combined with Supervised Machine Learning algorithm and natural language processing to add strength in the classification of, and then prediction for, text reviews. It describes the project implementation and design, including data collection and methodology. The chapter focuses on the results, comparing the accuracy and the performance of both FE algorithms combined with the variety of ML classification algorithms.
