Location identification for crime & disaster events by geoparsing Twitter

• Published in: International Conference for Convergence for Technology-2014 pp. 1 - 3
• Main Authors: Dhavase, Nikhil, Bagade, A. M.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.04.2014
• Subjects: Buildings; Conferences; Data mining; disaster response management; event classification; Event detection; geographic information retrieval; Geoparsing; Media; Real-time systems; Twitter
• DOI: 10.1109/I2CT.2014.7092336

Geoparsing means automatically identifying locations in text. The location mentions in messages during crime and disaster events are very crucial, as they can help emergency response teams to quickly identify the place to send rescue teams to the location. Use of social media during such crisis events has been rapidly increasing all over the world, as well as in India. We consider here the source of messages as Twitter because it is realtime, robust and can handle large amounts of data. We collect tweets at real time and then parse those tweets for crisis situation and location information. Extracting the location information to the level of streets & buildings will help to detect the exact location of the event; this is done with the help of NLP methods. We use classifiers to classify tweets to obtain the event occurred.