New Legal Database Uses Concept Mining to Reduce Duplication

Scientio LLC, creators of concept mining software have succesfully trialed their ConceptMap product with Legis Editores, the largest legal publishers in the Spanish-speaking world. ConceptMap finds duplicates and near duplicates in large databases of text efficiently using the concepts, rather than the words in a document.

Woburn Sands, UK (PRWEB) January 21, 2006 -- Legis Editores have a database of 2.5 million legal documents which contained exact duplicates, near duplicates, revised versions and various other forms of duplicated content. Through their Subsidiary, Iconomultimedia of Columbia they contacted Scientio to find a .Net solution to locating duplicates in the existing database, and whenever new content was added.

Scientio's ConceptMap product has succesfully solved this problem, and can find the 10 nearest example documents from the database to any new document in under 3 seconds.

ConceptMap has application in de-duplication and detecting plagiarism, as well as enabling new products like concept based search engines and concept based matching of similar web pages and other content. It is available in .Net and Java versions, and supports multiple languages, initially English and Spanish.

Scientio creates and sells component software making use of research in Artificial Intelligence, Computational intelligence, Fuzzy logic and Chaos theory. Scientio LLC has offices in West Palm Beach, USA and Woburn Sands in the UK.

# # #