Spatio-Temporal Decision Support System for Natural Crisis Management with TweetComP1

This paper discusses the design of a social media crisis mapping platform for decision making in natural disasters where tweets are analysed to achieve situational awareness during earthquake and tsunami events. A qualitative end user evaluation is undertaken on our first prototype system to get feedback from practitioners working in the field of hazard detection and early warning. Participating in our evaluation is the Kandilli Observatory and Earthquake Research Institute (KOERI) and the Portuguese Institute for the Sea and Atmosphere (IPMA). We conclude that social media crisis mapping is seen as a valuable data source by control room engineers, with update rates of 10-60 seconds and false positive rates of 10-20% (general public incident reports) needed. Filtering crisis maps and statistical reports by social media platform and user type is desirable as different report sources have different credibility