IIA Projects in 2012
Metadata Records Translation and Evaluation for Multilingual Information Access to Digital Collections
Digital collections contain digital objects in different formats to serve a defined community or a set of communities. Libraries and museums in the U.S. have developed numerous digital collections in order to preserve scientific, cultural, and heritage materials and to provide convenient access for their users by organizing the objects and representing them through metadata. However, most of these collections can only be accessed in English. Very few digital collections in the United States support multilingual information access (MLIA) that enables users to search, browse, recognize and use information from multilingual digital objects. Metadata Records Translation (MRT) is the process of converting metadata records describing objects in a digital collection from one language into other languages. It is the necessary first step to enable MLIA for a digital collection. Machine translation (MT), or strategies combining MT with human efforts, requires exploration as human translation of metadata is expensive and time-consuming.
This project aims to
1) evaluate the extent to which current machine translation technologies generate adequate translation for metadata records and
2) identify the most effective metadata records translation strategies for digital collections. The long-range result of implementing effective machine translation of metadata records will be to make digital collections more useable by a much more linguistically varied user population, and thus enabling broader access to information stored in digital collections.
Please visit the project website: http://max.lis.unt.edu/MRT/ for more information.