Prediction of Satisfaction from Facial Expression: Sentiment Analysis Approach

Abstract

 Digital Image Processing (DIP) is the process of digital images using various computer algorithms. This digital image processing has been employed in a number of areas such as pattern recognition, remote sensing, image-sharpening, colour and video processing and medical. Machine Learning techniques like CNN(Convolution Neural Networks) are found helpful in providing accurate prediction, in this paper there is a representation of an automated review system. Reviews are collected specially reviews in two forms as manual entries and video recordings. Both the reviews will be analysed and generalized comparison will be carried out. Basic idea in this paper is to identify the success rate of automatic review systems using facial expressions. It is  very essential in determining the satisfaction level of the person with respect to any service or facility provided . Video review captures most of the indirect expressions which is helpful in deciding the emotion through video analysis; much more accurate results are obtained. It can be used in several aspects like movie review, product review, student’s feedback, etc. Various technological aspects are touched in this innovative concept of sentiment analysis.