Sentiment Analysis of Different E-commerce platform reviews using Machine Learning Algorithm

Abstract

E-commerce provides different products/services to the customer where customers can easily get their desired products anywhere they want. While buying online, they rely on the product reviews made by other users which gives much more emphasis on the product review as it is required for the selection of a product. For the analysis of such reviews, sentiment analysis is done. Since the data are in huge numbers, machine learning algorithms are used for the fast and effective calculation and analysis of these product reviews. These reviews can be done quickly using machine learning as the model is created where thousands of reviews are made. For the better accuracy of the model, the datasets are subjected to different pre-processing techniques. Then, both supervised and unsupervised learning methods as well as the deep learning method to classify the sentiments of the dataset in positive or negative class. For the validation of our model, secondary dataset were obtained from the ecommerce platforms like Daraz. For supervised learning models, we have used Naive Bayes and SVM classifiers. From lexicon based analysis, VADER classifier is used which exhibits 68% accuracy when validating with the secondary data. Also supervised algorithms like SVM classifiers exhibit 71% accuracy whereas Naive-Bayes classifiers exhibit 68% accuracy for the data gathered. But the highest accuracy was obtained from deep learning models which exhibit 75% accuracy.