Sentiment Analysis of Amazon Reviews using NLTK Vader and Robert

Abstract

With the advancement of technology, huge volume of data is exchanged every-day, example - from social media sites like Instagram, Twitter, or e-commerce sites like Amazon, Flipkart etc. The human brain is unable to process this large amount of data (mainly out of which is unstructured) and this results in the development of technology which do the work of humans. One such methodology is Sentiment Analysis. Sentiment analysis plays a crucial role in understanding public opinion and consumer behavior. In this paper, we conduct a comparative study of sentiment analysis on Amazon product reviews using two different sentiment analysis approaches: VADER (Valence Aware Dictionary and Sentiment Reasoner) and RoBERTa (Robustly Optimized BERT Approach). We explore the effectiveness of these models in capturing the sentiment expressed in Amazon reviews, considering various aspects such as accuracy, computational efficiency, and generalization capabilities. Our findings provide insights into the strengths and limitations of each approach, contributing to the advancement of sentiment analysis methodologies. Keywords Sentiment Analysis, Roberta, Vader, Machine Learning.