Leveraging Cross-Lingual Tweets in Location Recognition

Abstract

The increased popularity of micro-blogging applications (e.g. Twitter) have resulted in the creation of large streams data. Such data provides a great opportunity for researchers to explore event detection. In particular, road traffic detection is of great importance to various applications, i.e. Intelligent Transportation Systems. Recognizing locations in the text of tweets plays an essential role in traffic detection. In this paper, we propose a novel method to identify locations in tweets using cross-lingual (English and Arabic) data collected from Twitter. The collected data (tweets) will be filtered to give emphasis to the United Arab Emirates, UAE, region. Then, features are extracted from the data to classify the tweets into traffic-reporting and non-reporting. The classified tweets are geoparsed and geocoded to acquire the location of reported traffic.