Real Time Detection of Traffic from Twitter Stream Analysis

The social network has been newly engaged as a source of information for event discovery, with particular mention to road traffic jamming and the car accident. In this thesis, We are going to present a real-time monitor system for traffic event detection from Twitter stream study. The system fetches tweets from Twitter according to some search criterion like processes tweets, by apply text mining techniques and finally perform the categorization of tweets. The aim is to allocate the appropriate class label to each tweet as related to a traffic incident or not. The traffic detection system was engaged for real-time monitoring of several areas of the Italian road network, allowing for detection of traffic events almost in real time, often before online traffic news web sites. We employed the support vector machine as a classification model and we achieved an accuracy value of 95.75% by solving a binary classification problem (traffic versus non-traffic tweets). We were also able to discriminate if traffic is caused by an external event or not, by solving a multiclass classification problem and obtain accuracy value of 88.89%.