Feature Extraction and Classification of Movie Reviews
Nhamo Mtetwa, Awukam Ojang Awukam, Mehdi Yousefi · 2018
Sentiment analysis identifies a user's attitude towards a service, a topic or an event and it is very useful for companies that receive many written reviews of their services. We investigate the effect of feature extraction techniques on supervised machine learning classifiers using four different performance metrics using a publicly available movie review dataset. Our objective is to explore different classification algorithms as well as utilizing diverse feature extractors and compare outcomes and finally select the trio of feature extraction technique, classification algorithm and performance metric with the best result for the movie review classification use case.