The Deceptive Language Processing Framework: Fusing Top-Down and Bottom-Up Approaches to Deception Discovery Navy STTR FY2010.A Sol No.: Navy STTR FY2010.A Topic No.: N10A-T029 Topic Title: The Deceptive Language Processing Framework: Fusing Top-Down and Bottom-Up Approaches to Deception Discovery Proposal No.: N10A-029-0533 Firm: Intelligent Automation, Inc. 15400 Calhoun Drive Suite 400 Rockville, Maryland 20855 Contact: Xiong Liu Phone: (301) 294-4629 Web Site: www.i-a-i.com Abstract: The exponential growth of text-based communication associated with the Internet has lead to a vast increase in the amount of unstructured messages that open source intelligence needs to process. This increase has lead to the need to develop methods for facilitating the detection of deception in various forms of text-based messages, from chat rooms, emails, weblogs, to text messaging. Methods are required to discover hidden messages, hostile disinformation, and author misrepresentation. To address the critical need of marrying theoretical and computational approaches to deception detection, Intelligent Automation, Inc. (IAI) proposes to develop a novel Deceptive Language Processing (DLP) framework for deception analysis of large-scale quantities of text. DLP synthesizes social and psychological theory with computational techniques (e.g., natural language processing, data mining) for modeling the relationships between discourse and deception in its various forms. Benefits: The project will lead to new approaches that will enhance our understanding of deception and improve our ability to detect digital forms of deception. This project will make available a suite of tools for the analysis of messages that will be useful for analyzing national security questions. With these tools the ability to predict group dynamics, affiliations, and deception on the basis of language and discourse profiles is quite plausible. In theory, our methods can provide very fast and efficient markers of basic social dynamics, and these tools will help analysts understand past actions and cognitions of previous regimes and assess emerging threats. In the commercial sector we anticipate that the resulting technology will have application in the corporate world and in the insurance industry. Return