Sentiment Analysis on Amazon Reviews Using NLTK

Sentiment Analysis on Amazon Reviews Using NLTK

Kalyan kumar Vaka, Rishitha Chirumamilla, Mrs Parimala Garnepudi, Rajeshwari Thammi

Abstract— Natural Language Processing (NLP) has gained significant interest in the area of sentiment analysis, which involves identifying subjective information like emotions, attitudes and opinions from textual data provided by customers. In this paper, we present a sentimental analysis approach for Amazon reviews using the Natural Language Toolkit (NLTK). We collected a dataset of Amazon reviews related to various products and used NLTK for preprocessing and analyzing the text data. The preprocessing steps included text normalization, tokenization, stop-word removal, and stemming. We then applied NLTK's built-in sentiment analysis algorithm to classify each review as positive, negative, or neutral. The results of our analysis showed that NLTK-based sentiment analysis achieved a high accuracy rate in classifying the sentiment of Amazon reviews. Moreover, we performed a detailed analysis of the positive and negative reviews, identifying the most frequently mentioned aspects of the products, food and the reasons behind the positive and negative sentiments. Our study highlights the effectiveness of the NLTK tool in sentiment analysis and its potential applications in the e-commerce domain for understanding customer opinions and improving product quality.

Keywords—NLTK (Natural Language Toolkit), Text classification, machine learning, tokenization, stop words, Naive based classifier, logistic regression.