Abstract
With increasing interest in e-commerce and online shopping recently, purchasing items online has grown to be more and more fashionable outstanding in options like lower in prices, and best quality products with high positive reviews, therefore customers are seeking to shop online. User reviews are more useful in deciding product quality. Customer reviews are one of the most important elements that determine customer satisfaction with the products. Moreover, it gives a better picture of products to business proprietors. Hence, this paper aims to conduct a sentimental analysis approach on a group of customer reviews collected from Flipkart. As well, classify each review into one of these classes: positive review or negative review by using Natural Language Processing (NLP). Online reviews have enough potential to provide a conclusion to the buyers about the product and its quality, performance, and recommendations, in this manner providing a detailed picture of the product to the end buyers. These online reviews are not only useful for customers but also used for manufacturers to realize customer requirements. Both positive reviews and negative reviews of customers play a major role in determining the requirements of customers and extracting customer feedback about the product. Sentiment analysis is an approach that helps to extract useful information, like opinions, attitudes, emotions, etc, from text data. It consists of different approaches, including Extraction, Tokenization, Lemmatization, and Classification. Extracting reviews from the website using the Web scraping package Beautiful Soup, we can easily fetch the brand name, reviews, ratings, and other related things for a product, using NLTK library that helps in classifying the reviews as positive review and negative review, and finally, they will come to know which product having more number of positive reviews and best reviews.