Published Paper Details:

Sentiment Analysis of the comment textual content from the social media is beneficial for understanding the general public opinion on the product review. The core of sentiment analysis is the text classification task, and distinct words have different contributions to classification. The classification provides that the product is positive or negative primarily based on the comment text provided by the customers of the product. Our proposed system uses the conventional TF-IDF algorithm and generates weighted word vectors. The weighted term vectors are given as input to the BiLSTM to capture the context information effectively. The sentiment is positive or negative of the comment is obtained by feed forward neural network classifier. The system will be tested with the comment text collected by the customer product review from social media, e-commerce site, and the result indicates that the product has the positive or negative reviews.