Integrated Feasibility Experiment for Bio-Security: IFE-Bio, A TIDES Demonstration

Integrated Feasibility Experiment for Bio-Security: IFE-Bio A TIDES Demonstration Lynette Hirschman, Kris Concepcion, Laurie Damianos, David Day, John Delmore, Lisa Ferro, John Griffith, John Henderson, Jeff Kurtz, Inderjeet Mani, Scott Mardis, Tom McEntee, Keith Miller, Beverly Nunan, Jay Ponte, Florence Reeder, Ben Wellner, George Wilson, Alex Yeh The MITRE Corporation Bedford, Massachusetts, USA and McLean, Virginia, USA 781-271-7789 lynette@mitre.org ABSTRACT As part of MITRE’s work under the DARPA TIDES (Translingual Information Detection, Extraction and Summarization) program, we are preparing a series of demonstrations to showcase the TIDES Integrated Feasibility Experiment on Bio-Security (IFE-Bio). The current demonstration illustrates some of the resources that can be made available to analysts tasked with monitoring infectious disease outbreaks and other biological threats. Keywords Translation, information extraction, summarization, topic detection and tracking, system integration. 1. INTRODUCTION The long-term goal of TIDES is to provide delivery of information on demand in real-time from live on-line sources. For IFE-Bio, the resources made available to the analyst include e- mail, news groups, digital library resources, and eventually (in later versions), topic-specific segments from broadcast news. Because of the emphasis on global monitoring, there is a need to process incoming information in multiple languages. The system must deliver the appropriate information content in the appropriate form and in the appropriate language (taken for now to be English). This means that the IFE-Bio system will have to deliver news stories, clusters of relevant documents, threaded discussions, alerts on new events, tables, summaries (particularly over document collections), answers to questions, graphs and geo- spatial temporal displays of information. The demonstration system for the Human Language Technology Conference in March 2001 represents an early stage of the full IFE-Bio system, with an emphasis on end-to-end processing. Future demonstrations will make use of MITRE’s Catalyst architecture, providing an efficient, scalable architecture to facilitate integration of multiple stages of linguistic processing. By June 2001, the IFE-Bio system will provide richer linguistic processing through the integration of modules contributed by other TIDES participants. By June 2002, the IFE-Bio system will include additional functionality, such as real-time broadcast news feeds, new machine translation components, support for question- answering, cross-language information retrieval, multi-document summarization, automatic extraction and normalization of temporal and spatial information, and automated geospatial and temporal displays. 2. The IFE-Bio System The current demonstration (March 2001) highlights the basic functionality required by an analyst, including: • Capture of sources, including e-mail, digital library material, news groups, and web-based resources; • Categorizing of the sources into multiple orthogonal hierarchies useful to the analyst, e.g., disease, region, news source, language; • Processing of the information through various stages, including “zoning” of the text to select the relevant portions for processing; named entity detection, event detection, extraction of temporal information, summarization, and translation from Spanish, Portuguese, and Chinese into English; • Access to the information through use of any mail and news group reader, which allows the analyst to organize, save, and share the information in a familiar, readily accessible environment; • Display of the information in alternate …

Content truncated to 500 words