Toward a Sentiment Analysis Framework for Social Media

Nowadays, opinions and sentiments can be easily expressed through social media and have a strong social impact. Thus, the need for an automated way to analyze the generated data with less human effort and more accuracy. In this respect, sentiment analysis tasks such as; preprocessing, classification, etc. provides various techniques that achieves notable accuracy scores, but presents limitations depending on the experimental context. Through our literature review, only few studies focused on establishing a reference framework for sentiment analysis. In this paper, we provide a literature review for common sentiment analysis tasks with discussion about future research trends, then we propose an abstraction model of a generic framework architecture for sentiment analysis in the context of social media based on previous works and enhanced with new concepts.