Sentiment classification and aspect-based sentiment analysis on yelp reviews using deep learning and word embeddings

Abstract

Opinion mining has significantly supported knowledge and decision-making. This research analysed the content of online reviews including the text of reviews and their rankings. The restaurant reviews of Yelp website have been analysed into two sentiment classifications, binary classification (positive and negative) and ternary classification (positive, negative, and neutral). Three different types of predictive models have been applied including: machine learning, deep learning and transfer learning models. In addition, we propose a new unsupervised approach to apply for aspect-level sentiment classification based on semantic similarity, which allows our framework to leverage the powerful capacity of pre-trained language models like GloVe and eliminated many of the complications associated with the supervised learning models. Food, service, ambience, and price are the aspects that have been categorized according to their sentiment context. In conclusion, 98.30% of the maximum accuracy obtained using ALBERT model. The proposed aspect extraction method achieved an accuracy of 83.04%.