Social networks have recently come across as a great source of information for detecting congestion, accidents, as well as crowding due to numerous festivals. Twitter is one of the most popular sites because it expresses knowledgeable information in minimum words. Since the tweets have limited words, processing it becomes easy. Hence, This project uses Twitter as a source of information. This project focuses on classifying whether a tweet is a traffic related tweet or not using Semantic technologies. In this project, the system fetches the tweets using the Twitter Api and then pre-process it, converting into tokens and cleaning all noise. These tokens are then semantically processed and classification is performed using Na¨ıve Bayes multinomial classifier. Different types of tweets data are used for prediction, including tweets from selected road-traffic Twitter accounts, tweets that contain road-trafficrelated keywords and geo-tagged tweets. These classified results are then plotted as a colored path on the android application. Such information can help the traveler to make a better travel plan.

Semantic Analysis geo-tagged tweets Na¨ıve Bayes multinomial classifier Tokenization Stop-word filtering Stemming Stem filtering Feature representation