Leveraging unscheduled event prediction through mining scheduled event tweets

A considerable portion of social media messages is devoted to current events. Aside from references toevents that recently happened, social media messages may also refer to events that have not occurred yet.Future events, such as football matches in the case study we present here, may be scheduled and knownto happen; other future events, such as transfers of football players, may only be rumoured, and may infact not happen in the end. We describe a news mining component that learns to identify tweets referringto scheduled and unscheduled future events, by being trained on messages referring to scheduled futureevents (as the latter are easy to harvest). Our results show that discriminating between tweets that referto upcoming football matches and tweets that refer to past matches can be done relatively reliably withsupervised machine learning methods. However, when these trained models are applied to unscheduledevents, performance drops to near-baseline performance. We discuss how these results can be explainedby the distinction between event type and event domain.