Role of Hybrid Optimization in Improving Performance of Sentiment Classification System

Researchers are employing sentiment analysis on user comments and tweets to get a feel for popular opinion. The rate at which the world is changing as a consequence of technological progress is unprecedented. In today's world, having access to the Internet is a need. To share their thoughts on current events, more and more people are turning to social networking apps. Collecting and analysing client feedback is crucial whether marketing a product or attempting to enhance a government service. Opinion mining, also known as sentiment analysis, is often done in advance of a discussion in which the attitudes behind different points of view are to be found. Sentiment analysis's popularity as a tool for understanding the public's mood has skyrocketed in recent years. In the newest research, neural networks are used to analyse the tweets of Twitter users. Researchers are increasingly interested in "tweets" (comments) and the substance of these expressions, and the use of Twitter data for survey research has increased in latest years. Purpose of this research is to assess performance of several sentiment analysis methods when applied to Twitter data. Present research has used compression mechanism to reduce image size and noise removal mechanism to improve the quality of image before applying CNN model. Thus integration of compression, noise removal and CNN results in hybrid model that is providing more accurate solution with better performance.