This is a simple tool to see if one can find a corelation between the sentiment of a person and the time so that an appropriate addvertisement could be targeted to them at that time. Phase 1: This involves getting tweets from twitter api, removing noise and stroing them in some semi structured form. Then processing the tweets.
Phase 2: This involves developing Naive bayes text classifier, random forest and SVM classifier for tweets.
Phase 3: Analyisng the result to mine some pattern with time using clustering algorithms.