Location-based event search in social texts

Abstract

A tremendous amount of information is being shared everyday on social media sites such as Facebook, Twitter, or Google+. Often times these social media texts include information that are valuable to others, such as activities (e.g., art fairs, jazz festivals, and gatherings), natural disaster occurrences (e.g., tornadoes, earthquakes), or incidents (e.g., traffic jams). However, current search on social media data is mostly keyword or hashtag based. The keyword based search method does not allow efficient search of events. In order to detect, scan, and search location based events from social media, users and social texts need to be effectively geotagged. Based on geotagged users and texts, effective and efficient algorithms need to be developed to integrate key word and spatial search to allow event query. This position paper summarize the current work, challenges, and research needs in location-event search in social texts.