Intelligent Classification and Clustering Techniques for Text Data Mining

Award Information
Agency:
Department of Defense
Branch:
Army
Amount:
$728,340.00
Award Year:
2002
Program:
SBIR
Phase:
Phase II
Contract:
DAAH01-02-C-R021
Agency Tracking Number:
A002-1399
Solicitation Year:
N/A
Solicitation Topic Code:
N/A
Solicitation Number:
N/A
Small Business Information
Scientific Systems Co., Inc.
500 West Cummings Park, Suite 3000, Woburn, MA, 01801
Hubzone Owned:
N
Socially and Economically Disadvantaged:
N
Woman Owned:
N
Duns:
859244204
Principal Investigator
 Sai-Ming Li
 Research Engineer
 (781) 933-5355
 eliot@ssci.com
Business Contact
 Raman Mehra
Title: President/CEO
Phone: (781) 933-5355
Email: rkm@ssci.com
Research Institution
N/A
Abstract
"This SBIR effort will develop an integrated information classification anddocument management system, applicable to complex weapons systems software.Currently, software engineers at Army's Tank-automotive & Armaments Commandrely on Software Trouble Reports (STRs) that contain unstructured text describing operational problems filed by soldiers fortroubleshooting of computer-controlled weapons systems.Past STRs and maintenance records provide a valuable source ofinformation that can help software engineers to understand newproblems, identify the faulty modules, and eventually provide valuableguidance on how to fix the problem.The overall objective of the Phase II effort is to develop aprototype Software Report Management System (SRMS) that will automaticallymanage STRs and associated maintenance records,extract useful information from the document archive, and discoverpreviously unknown domain knowledge that will assist maintenance of the system.It will also facilitate focused and accurate search forproblems/solutions/case-studies.To achieve the above objective, we propose to develop advancedclustering, information extraction, and data fusion algorithmsfor the document collection using textual analysis and machine learningtechniques. Such algorithms will be used to group the STRsinto meaningful clusters and extract useful information fromthem to build a knowledge base for software problems. We willthen integrate these algorithms in

* information listed above is at the time of submission.

Agency Micro-sites

US Flag An Official Website of the United States Government