Twitter Bot For Consumer Feedback Analysis

This project addresses the problem of analysing consumer reviews on Twitter. In other words, it separates tweets based on the feedback show them in good, bad or neutral. When it comes to top social networking website, twitter is one of them which allows their users to use this platform to put status of using 140 characters. With over 200 million users, it is a resource that is rapidly expanding, half of whom log in daily and generate about 500 million tweets per day. With such a high density of use, we hope that our analysis of the opinions expressed in the tweets will reflect public opinion. Analysis of public opinion is important for many applications. B. Companies that want to know how their products will react in the market, predictions of political elections, predictions of socio-economic phenomena such as the stock market. This project aims to develop a feature classifier that accurately and automatically classifies opinions in unknown tweet streams.