Cricket Twitter Data Sentiment Analysis and Prediction Exerted Machine Learning

The aim of this project is to create an algorithm that can accurately classify Twitter messages as positive or negative, in relation to a query term. Our hypothesis is that to have high accuracy in separating emotions in Twitter messages using machine learning methods. We propose an approach that analyses feeling of cricket fans and correlate sentiment to match play. We use data collected on twitter in the message form. To predict the outcome of a cricket match we are not going to rely on a single machine learning algorithm we are using at least two machine learning algorithms to compare the accuracy. We have applied modern classification techniques -Logistics Regression and Random Forest, and conducted a comparative study based on the overall cricket tweets. The project outcome is given in form of webpage giving both analysis and prediction of live tweets using Logistic Regression and older tweets using Random Forest.