IDENTIFICATION OF PRACTICAL TRAFFIC VIA DIGITAL MEDIA TWITTER STREAM AND SCRUTINY

In the recent times, social networks have been extensively used as a data source for the event detection. Social networks permit people to generate an identity and allow them share it to construct a community. The resultant social network is a basis for managing of social relationships, discovering users with related interests, and locates content and knowledge entered by several users. We provide an actual monitoring scheme for traffic event recognition from the analysis of Twitter stream. The system was designed from ground as event-driven infrastructure, built on service oriented architecture and obtains tweets from Twitter based on various search criteria such as processes tweets, by application of text mining methods; and performs Tweet classification. The objective is to allocate the suitable class label to every tweet, as associated to traffic event or else not. The traffic detection system was in use for monitoring of numerous areas, allowing for recognition of traffic events more or less in real time, often prior to online web sites