Humanitarian Assistance/Disaster Relief (HA/DR) Social Media Analytics Tools

The staggering velocity of digital data creation combined with a shortage of human analytical resources creates a significant demand for automated systems that can efficiently discover, process, and visually display information in an organized manner. Using NLP technology to enable automated discovery, process, and displaying actionable information is the premise of this research topic"s technical approach. The primary capability developed for the first builds of Social Media Analytics Toolkit (SMAT) under the Phase I effort was the automated discovery of real-world disaster related events, and scoring and categorizing them into associated impact categories such as communications, fire, health hazard and infrastructure. The benefits of SMAT"s NLP-based analytics demonstrated during the Phase I effort were to: assist the development of filters; and to quickly drill-down and focus on different impact categories of a disaster. The Phase II plan is to further SMAT development, comprising task groups for: additional requirements collection and refinement, including relevant ones from the IV2 JCTD Operational Requirements, improving capability and performance, incorporating a distributed collaboration environment, componentization and integration into other tools and environments such as TweetTracker and CRAFT, and evaluation verification and validation through participation in real-world HA/DR and experimentation events such as the recent SMT-LOE.