|NetOwl NameMatcher wins MITRE Challenge|
SRA’s Name Matching Product Declared Best Among 40 Worldwide Competitors
FAIRFAX, Va., Oct. 13, 2011 — SRA International, Inc., a leading provider of technology and strategic consulting services and solutions to government organizations and commercial clients, today announced that its NetOwl® NameMatcher product recently won the MITRE Challenge™ for Multicultural Name Matching.
The MITRE Challenge™ is an open competition modeled after the Netflix Prize and encourages innovation in technologies of interest to the federal government. It is sponsored by the MITRE Corporation, a not-for-profit corporation chartered to work in the public interest (www.mitre.org). Forty teams from leading commercial companies and academic institutions around the world participated in the Multicultural Name Matching Challenge, which involved multicultural person name matching, a technology whose uses include vetting persons against a watchlist (for screening, credentialing and other purposes) and merging or de-duplication of records in databases. The top ten participants were invited to a Technical Exchange Meeting held at MITRE in McLean, Va., on Oct. 5, 2011, where the results were shared and the winner officially announced. SRA’s NetOwl was entered under the pseudonym “Mean Mr. Teach.”
“We are very pleased that our NetOwl NameMatcher product achieved the highest accuracy on the Challenge data and won this exciting competition,” said SRA Vice President and Director, Text and Entity Analytics, Dr. Chinatsu Aone. “Multicultural name matching is critical to both government and commercial clients who address such important issues as counter-terrorism, regulatory compliance, and risk management as well as anti-fraud, waste and abuse. NetOwl offers state-of-the-art solutions for all of these areas.”
NetOwl NameMatcher provides highly accurate and scalable name matching capabilities for a diverse collection of names spanning many cultures, ethnicities and languages. NameMatcher applies a unique machine learning approach to this significant name variant challenge by exploiting a large amount of empirical data. It automatically learns and generates tens of thousands of name-matching rules, avoiding the bottleneck of traditional approaches where manual development and tuning of rules are labor-intensive. NameMatcher significantly reduces the total cost of deploying and maintaining name matching solutions for clients while attaining the highest accuracy.
About SRA International, Inc.