SRA PARTICIPATION IN TIPSTER PHASE II 
POC: Lisa Rau 
SRA International, Inc. 
4300 Fair Lakes Court 
Fairfax, VA 22033 
lisa_rau@sra.com 
(703) 803-1851 
I. INTRODUCTION 
SRA, although not a research contractor under 
Tipster Phase II, nonetheless actively participated in 
the program in a variety of ways. Some of the 
activities include: 
Preparation and distribution of the Named 
Entity Tagging and Discourse Tagging 
Tools for theSixth Message Understanding 
Conference (MUC6) and Multilingual Entity 
Task (MET) 
Implementation / Award of two Tipster 
application prototype systems: Bluesky and 
USACOM 
Sole contractor on two Tipster-affiliated 
systems, the Intelligence Analysts Associate 
(IAA) under funding from Rome 
Laboratories for NAIC, and the Overture 
program for OIR. 
Participation in the Contractor Architecture 
Working Group (CAWG) 
Participation in the Common Pattern 
Specification Language subgroup 
Our current activities include support for the data 
extraction components of the ISX-led ISLE/InfoTech 
Tipster application. 
II. RESEARCH PROGRAM 
HIGHLIGHTS 
The application of machine learning 
techniques to reduce the customization time 
for data extraction systems 
The use of multilingual data extraction for 
targeted machine translation or gisting 
Fusion of data extraction technology with 
other media (speech; video) for multimedia 
fusion applications 
lII. EVALUATIONS 
SRA participated in the Sixth Message 
Understanding Conference in three of the four areas 
(Named Entity, Template Element, and Scenario 
Template). SRA participated in the Multilingual 
Entity Task (MET) in both Japanese and Spanish. 
IV. PHASE III PLANS 
On the research front, as a Tipster Phase III 
contractor, SRA intends to make significant advances 
in the customizability of extraction technology, 
primarily by bringing together our expertise in 
machine learning and natural language processing. 
A second focus of our Phase III research plans 
involve the intelligent summarization of texts. 
Finally, we intend to continue to contribute to the 
Contractor Architecture Working Group and other 
common efforts, and complete the Tipster application 
systems. 
SRA engages in an active program of research and 
operational support in both multilingual data 
extraction, natural language processing more 
generally, and text retrieval integration. Our current 
research thrusts involve: 
53 
