Sentiment analysis has evolved over the past few decades, Sentiment analysis has evolved over the past few decades. However, most of the work in it revolved around textual sentiment analysis. On the other hand, audio-visual sentiment analysis is still in a nascent stage. It is hard to identify an emotion based on only facial expression as it can cause ambiguity since expressions of a human face may not convey the exact emotion as their speech does. Identifying sentiments is mostly hindered by 2 classes (positive, negative) or in some cases 3 classes (positive, negative, neutral)

Keywords : Natural Language Processing, Sentiment Analysis, Neural Network, Image Processing.