An Automatic Participant Detection Framework for Event Tracking on Twitter

Topic Detection and Tracking (TDT) on Twitter emulates human identifying developments in events from a stream of tweets, but while event participants are important for humans to understand what happens during events, machines have no knowledge them. Our evaluation football matches basketball games shows that tweets is difficult problem exacerbated by Twitter’s noise bias. As result, traditional Named Entity Recognition (NER) approaches struggle identify the pre-event stream. To overcome these challenges, we describe Automatic Participant (APD) detect an event’s before starts improve machine understanding events. We propose six-step framework present our implementation, which combines information Wikipedia. In spite difficulties associated with NER challenging context approach manages restrict consistently detects majority participants. By empowering some about APD lays foundation not just improved TDT systems, also future where can model mine themselves.