E.g., 11/22/2019
E.g., 11/22/2019

Kantan BuildAnalytics™ Enhanced with Gap Analysis Technology

KantanMT


Friday, 24 January, 2014

KantanMT today announced the second release of its training data analytics suite – Kantan BuildAnalytics™. The release includes the launch of the new Gap Analysis technology which assists in the development of production ready KantanMT engines. Created as an integral part of Kantan BuildAnalytics™, Gap Analysis offers members a unique insight into their Machine Translation (MT) training material – helping them to build higher performing KantanMT engines in a shorter period of time.

Kantan BuildAnalytics includes several technologies; the Training Rejects report identifies segments that are rejected by the KantanMT data cleansers; the BLEU, TER and F-Measure distribution reports provide a detailed analysis into how an engine is performing, and can be used when fine-tuning training data; and The Gap Analysis detects unknown or untranslated words (data gaps) which can then be filled using relevant training material or terminology data.

“Most quality improvements for SMT systems will be made by fine tuning training data and filling data gaps,” said Tony O’Dowd, Founder and Chief Architect, KantanMT. “Post-editing raw-MT output and focussing on minimizing data gaps will significantly improve the quality performance of a KantanMT engine.”

“KantanMT BuildAnalytics technology gives KantanMT members complete transparency when training engines,” said Eric Chubb, Senior Software Developer, KantanMT. “It offers our members invaluable insights into their training data, and empowers them to develop and maintain superior data sets which will produce even higher performing MT engines.”

For more information about Kantan BuildAnlaytics or to register for a demo, please contact Niamh Lacy ([email protected]).

What is KantanMT™?

KantanMT is a cloud-based implementation of Moses Statistical Machine Translation technology. Leveraging the power and flexibility of the cloud, KantanMT effortlessly scales to generate a high-quality, low-cost Machine Translation platform for organizations and small-to-medium-sized localization service providers. According to the research firm, Markets and Markets, the natural language processing sector will grow from €2.8 billion in 2013 to €7.4 billion in 2018, representing a compound annual growth rate of 21.1 percent.