Crowdsourcing using Intelligent Supervision to address Information Requirements in Crisis Situations (CRISIS)

Marine Expeditionary Units (MEUs) are often first responders, addressing crises such as natural disasters and regional instability. To adequately respond, they must quickly and accurately analyze large amounts of raw information. Fortunately, through crowdsourcing, there are vast, knowledgeable, and unexploited resources in local and military populations that can address information requirements (IR) and analysis needs. However, Marines need an effective system to use these distributed human processing resources. MapReduce addresses a similar distributed parallel computational processing problem, but not the inherent diversity of human resources. Inspired by MapReduce, we propose to design and demonstrate a system for Crowdsourcing with Intelligent Supervision to address IRs in Crisis Situations (CRISIS). CRISIS provides a crowdsourcing algorithm to employ diversely skilled crowds to address problems for MEUs, focusing on four key components: (1) an evolvable ontology of crowd capabilities to drive human-centered problem partitioning; (2) probabilistic models of crowd skills and preferences to not only identify crowd resources, but understand their capabilities; (3) a market-based optimization system to perform load management and mapping while addressing a scarcity of properly skilled resources to address tasks; and (4) a combination of semantic reasoning and provenance modeling to reduce solutions, identifying consistencies and inconsistencies in the process.