Traffic Event Related Blog Post Classification by Using Traffic Related Named Entities

Real-time monitoring of traffic flow requires physical sensors to be deployed on road networks. Development of such systems might be impractical due to deployment costs of sensors on large scale networks. This study presents a method to extract traffic event related tweets from social streams in order to employ users of social media as human sensors of traffic conditions or events. The proposed method offers a cost effective way of monitoring events or conditions affecting traffic flow. The method consists of three steps. The first step involves natural language processing tasks for preprocessing the blog posts. The second step extracts a set of traffic event related named entities from blog post texts using the model that is constructed with Conditional Random Fields. The third step includes classification in order to detect blog posts reporting events or conditions affecting traffic flow. The proposed method is experimentally evaluated on a set of tweets collected in one month under varying feature sets. The results show the potential of the approach for traffic monitoring and reveals that the use of traffic related named entities increases the classification accuracy.