Social Media Location Mining

Social media has become so ubiquitous and pervasive in our society that almost all organizations nowadays use it for various types of strategic objectives such as marketing, campaigns, publicity, information dissemination and information gathering. This has resulted in a flurry of research focused specially on processing social media texts because of the richness and the range of information that is embedded in them. Tweeter is a social media platform which allows 140 character messages to be posted for general as well as closed group communication. It is used extensively to upload immediate news about events by users, hence has an enormous potential to be used as up to the minute information source. Among the various types of information that can be mined from Twitter messages, location mining has attracted attention because of its application in identifying the geographical location of a topic in the Tweeter messages. For example tweets can be used to extract information on the location of an accident, a natural disaster, a concert and more recently, open invitation parties gone out of hand.

As part of this research we aim to develop models that would be able to extract location information from short Tweet type blogging texts.