Landslide Information Service by Integrating Multiple Information Sources

Abstract

Detection of natural disasters mostly depends on physical sensors, but few physical sensors are available for the detection of multi-hazards, such as landslides (Musaev, Wang & Pu, 2015). As a popular platform for real time information with users all around the world, social media augments the traditional way of natural disaster detection by providing supplemental timely data. LITMUS (Landslide Detection by Integrating Multiple Sources) is the application we built to provide landslide detection service that combines data from both physical and social information services by filtering and then joining the information flow from those services based on their spatiotemporal features (Musaev, Wang & Pu, 2015). In this thesis, we have explored additional information sources (news source and multilingual information from Twitter), which, if integrated to LIMTUS, can potentially increase credibility and coverage of the application.