Sentiment analysis of customer comments of e- commerce site applications with a machine learning approach

In this thesis, it is aimed to design a system that extracts summary information from natural language product or service reviews in order to benefit both users and producers of e-commerce platforms. The findings obtained by performing sentiment analysis of the comments on the Google Colab platform using Python language based on the comments of the users obtained by web scraping method over the Google Play market of three leading e-commerce companies in Turkey are presented. There are millions of reviews of these applications on Google Play. These reviews provide valuable information to the makers of the application and the user. These can include positive and negative comments about the company, cargo complaints, product satisfaction and even application errors. All the data obtained within the scope and as a result of the thesis study were analyzed with sentiment analysis and machine learning methods and the results were compared in terms of applications.