Translation memory systems containing a database of previously translated materials are widely used by translators. An important part of these systems is the matching algorithm that determines what translations get retrieved from the bank of available translations to assist the translator. A fundamentally important factor that determines the usefulness of a translation memory technology is how well the bank of previously translated materials matches up with the materials to be processed. It is also necessary to have technology powerful enough to manage a large amount of data.
Researchers at the University of Maryland’s Center for the Advanced Study of Language developed an algorithm for computer-aided translation software that produces more helpful translation examples by enabling translation memory systems to identify more relevant matches. This technology is consistently judged to be most helpful by translators for multiple domains and language pairs.
Applications· Computer-aided translation software
· Translation memory systems
Advantages· Produces more helpful translation examples
· Offers ability to handle larger amounts of data for translation memory systems
Office of Technology Commercialization
2130 Mitchell Building
7999 Regents Dr.
University of Maryland
College Park, MD 20742
Phone: (301) 405-3947 | Fax: (301) 314-9502