Basis Technology Places in Top Tier of MITRE Name Matching Software Challenge
Rosette Name Indexer earns high honors across all metrics and test sets for multicultural name matching
Cambridge—October 11, 2011—Basis Technology today announced its top tier finish in the MITRE Multicultural Person Name Matching Challenge in which 40 teams competed over eight months to accurately find all matching names within a given data set.
During the contest, participants submitted multiple results from their software, with a scoreboard displaying current standings as new submissions were graded. Throughout the challenge, Basis Technology submitted results from its Rosette® Name Indexer product rather than write custom code fitted to the challenge as some other teams did.
From its first submission, Rosette placed in the top tier of competitors, and stayed there through to the final judging last month across all metrics and test sets—beating out some very large company competitors. Rosette also offers the most comprehensive language coverage of the top competitors, matching names in English, Arabic, Chinese, Japanese, Korean, Pashto, Persian (Farsi/Dari), Russian, and Urdu.
“We saw the MITRE Challenge as an opportunity both to prove and to improve our name matching product,” said David Murgatroyd, Director of Engineering and leader of the Basis Technology team. “We entered the competition with Rosette as it was off-the-shelf, and focused on improving it, rather than writing custom software exclusively tailored to the challenge’s test data. Thus, our customers can be confident that they are receiving the same level of accuracy that put us in the top tier.”
The challenge was an innovative approach to MITRE’s review of identity matching and resolution tools for use by government sponsors. Name matching is critical for tasks such as verifying eligibility for Social Security or medical benefits, identifying and reunifying families in disaster-relief operations, and vetting persons against a travel watch list.
The improved Rosette Name Indexer version 4.2 that earned a top tier position in the MITRE Challenge™ is available now. Contact Basis Technology to request a free 30-day evaluation of Rosette.
About Rosette Name Indexer
Rosette Name Indexer uses fuzzy matching to find personal, location, and organizational names. It matches names across spelling variations and errors, different languages, nicknames, initials, mixed word order, missing name components, and other variations. The name matcher is very accurate, highly scalable, and easily integrated with applications and databases through a Java API or web service interface. It is currently in use by the U.S. intelligence community and financial institutions.
About Basis Technology
Basis Technology develops innovative products and solutions incorporating multilingual text analytics and digital forensics. Our Rosette® linguistics platform provides morphological analysis, entity extraction, name matching, name translation, and Arabic chat translation, yielding useful information from unstructured data in such fields as information retrieval, government intelligence, e‑discovery, and financial compliance. Our digital forensics team pioneers better, faster, and cheaper techniques to extract forensic evidence, keeping government and law enforcement ahead of exponential growth of data storage volumes.
Our products and services are used by over 250 major organizations, including Amazon.com, EMC, Endeca, Exalead/Dassault, Fujitsu, Google, Hewlett-Packard, Microsoft, Oracle, and governments around the world. Learn more at www.basistech.com or call +1-617-386-2090.