Improving National and Homeland Security through a proposed Laboratory for Information Globalization and Harmonization Technologies (LIGHT)

Abstract

A recent National Research Council study found that: "Although there are many private and public databases that contain information potentially relevant to counter terrorism programs, they lack the necessary context definitions (i.e., metadata) and access tools to enable interoperation with other databases and the extraction of meaningful and timely information" [NRC02, p.304, emphasis added] That sentence succinctly describes the objectives of this project. Improved access and use of information are essential to better identify and anticipate threats, protect against and respond to threats, and enhance national and homeland security (NHS), as well as other national priority areas, such as Economic Prosperity and a Vibrant Civil Society (ECS) and Advances in Science and Engineering (ASE). This project focuses on the creation and contributions of a Laboratory for Information Globalization and Harmonization Technologies (LIGHT) with two interrelated goals:

(1) Theory and Technologies: To research, design, develop, test, and implement theory and technologies for improving the reliability, quality, and responsiveness of automated mechanisms for reasoning and resolving semantic differences that hinder the rapid and effective integration (int) of systems and data (dmc) across multiple autonomous sources, and the use of that information by public and private agencies involved in national and homeland security and the other national priority areas involving complex and interdependent social systems (soc).

This work builds on our research on the COntext INterchange (COIN) project, which focused on the integration of diverse distributed heterogeneous information sources using ontologies, databases, context mediation algorithms, and wrapper technologies to overcome information representational conflicts. The COIN approach makes it substantially easier and more transparent for individual receivers (e.g., applications, users) to access and exploit distributed sources. Receivers specify their desired context to reduce ambiguities in the interpretation of information coming from heterogeneous sources. This approach significantly reduces the overhead involved in the integration of multiple sources, improves data quality, increases the speed of integration, and simplifies maintenance in an environment of changing source and receiver context - which will lead to an effective and novel distributed information grid infrastructure. This research also builds on our Global System for Sustainable Development (GSSD), an Internet platform for information generation, provision, and integration of multiple domains, regions, languages, and epistemologies relevant to international relations and national security.

(2) National Priority Studies: To experiment with and test the developed theory and technologies on practical problems of data integration in national priority areas. Particular focus will be on national and homeland security, including data sources about conflict and war, modes of instability and threat, international and regional demographic, economic, and military statistics, money flows, and contextualizing terrorism defense and response.

Although LIGHT will leverage the results of our successful prior research projects, this will be the first research effort to simultaneously and effectively address ontological and temporal information conflicts as well as dramatically enhance information quality. Addressing problems of national priorities in such rapidly changing complex environments requires extraction of observations from disparate sources, using different interpretations, at different points in times, for different purposes, with different biases, and for a wide range of different uses and users. This research will focus on integrating information both over individual domains and across multiple domains. Another innovation is the concept and implementation of Collaborative Domain Spaces (CDS), within which applications in a common domain can share, analyze, modify, and develop information. Applications also can span multiple domains via Linked CDSs. The PIs have considerable experience with these research areas and the organization and management of such large scale international and diverse research projects.

The PIs come from three different Schools at MIT: Management, Engineering, and Humanities, Arts & Social Sciences. The faculty and graduate students come from about a dozen nationalities and diverse ethnic, racial, and religious backgrounds. The currently identified external collaborators come from over 20 different organizations and many different countries, industrial as well as developing. Specific efforts are proposed to engage even more women, underrepresented minorities, and persons with disabilities.

The anticipated results apply to any complex domain that relies on heterogeneous distributed data to address and resolve compelling problems. This initiative is supported by international collaborators from (a) scientific and research institutions, (b) business and industry, and (c) national and international agencies. Research products include: a System for Harmonized Information Processing (SHIP), a software platform, and diverse applications in research and education which are anticipated to significantly impact the way complex organizations, and society in general, understand and manage critical challenges in NHS, ECS, and ASE.