The THOR (Text-enabled Humanitarian Operations in Real-time) framework, funded under the DARPA LORELEI program, is designed to provide humanitarian and disaster relief planners superior situation awareness through visual and analytical data science. The framework is a collaborative effort between industrial and university research laboratories (e.g., Next Century Corporation, ISI, RPI), and uses open-source technology with an intent to support the Intelligence Community, as well as military and civilian disaster humanitarian relief operations. THOR presents a unique application of data science methods for an important social purpose. At its core, THOR is powered by a domain-specific knowledge graph, which is derived from natural language outputs, and on which further analyses (e.g. event entity resolution) can be conducted. THOR is designed for low-resource linguistic environments, heterogeneous data sources and types (including social media and intelligence report fragments), arbitrary disasters that may not be known in advance, and the requirement of advanced graphical interactions.