When making purchasing or viewing decisions, reviews can be influential. However, traditional sentiment analysis of reviews relies solely on the words used by reviewers, which can lead to confusion when ambiguous language is employed. To accurately identify and classify positive and negative sentiment in reviews, a more sophisticated sentiment analysis model is necessary. Our proposed model improves upon existing methods by incorporating ratings into the decision-making process of CNN and LSTM deep-learning models, resulting in higher accuracy and reduced occurrences of false positives and false negatives. By taking ratings into account, our model is able to mitigate the impact of ambiguous language in reviews, making our analysis more accurate and efficient.

CNN, LSTM,Sentiment classification, ratings, model