Lexical Link Analysis for the Haiti earthquake Relief Operation Using Open Data Sources

Abstract

In the wake of the Haiti earthquake disaster, civil and military organizations engaged in vigorous relief operations to achieve rapid deployment of logistics, transport, security and medical supplies. Organizations involved in the operation collaborated with the expectation of effectiveness. To know that operations were effective - or understanding how to improve their effectiveness -- would require sifting through large volumes of communicatin to understand how verious oganizations collaborated and to improve future operations. in this paper, we will report how to leverage "Lexical Link Analysis" (LLA), as one means to achieve "System self-awareness" (SSA), and develop this as a durable methodology. We consider that the cognitive interface between decision makers and a complex system, (e.g., the organizational behaviors in an interagency operation) may be exprressed rhrough lexical link features embadded in documents and communications. The LLA method is composed of lexical analysis and link analysis. Lexical analysis is a form of link mining in which word pairs are extracted from document sets. Link analysis is a network analysis that explores key relationships among objects. We will report how to combine both the LLA and SSA methods to sift through real-life, open-source data, applied in this example to Haiti relief operations.