An E-Commerce Product Feedback Review Using Sentimental Analysis

Abstract

Research in natural language processing and text mining has been active in this area for the past few years. Many companies are using a sentimental analysis approach to get feedback from customers. So, the main objective is to use the result of this analysis to understand and analyse the feedback given by the customers. A website will be used in this study to collect information for sentimental analysis where feedback has been taken from customers by using the website. After, customers fill out the feedback forms. The backend data of customers is stored in MySQL. Data from the website is stored as a CSV file and the CSV file is used to get the analysis result. In this way, an analysis can be done regarding the input given by users, whether they have given the right feedback, and if the product is valuable or not. In this way, an idea regarding customers' way of giving feedback can be analysed. The main challenge is collection of the data for the study. For analysis, many solutions use Naïve Bayes. The main disadvantage is that, here in naive bayes, an assumption has been made based on the probabilities of the logs. In the Naive Bayes rule, the position of words is not considered, so the sentiment cannot be predicted based on the context of a document. The goal of this study is to review previous studies of analysis and examine the best method to compare customer feedback.