Analyzing Online Reviews of Customers Using Machine Learning Techniques

Abstract

Web data mining can provide insight into the e-commerce data which will be useful to promote sales, understand customers, and support new business opportunities. But understanding the customers and increasing their participation in online shopping is a great challenge as it requires the analysis of the relationship between customer satisfaction and the most significant factors that influence their online buying decisions. Machine learning comes into play in addressing the challenge faced by the e-commerce companies. In a nutshell, this paper focuses on the analytics of customers’ attitude towards using online shopping and intention to buy, using machine learning classifiers namely Naive Bayes, Logistic regression, Support Vector Machine, and Neural Network. For data analysis, Latent semantic analysis is applied to examine the most frequent words used in the online reviews. Finally, customer’s interest in online shopping analysis has been performed using machine learning classifiers to classify the customers’ sentiment from their posted reviews on the e-commerce platform. Also, we compared the performance results of these classifiers on the ecommerce dataset. The results reveal that the Naive Bayes classifier has performed better than all the other three classifiers in terms of execution time and the measures like accuracy, recall, F1-score, confusion matrix.