TMCnet - World's Largest Communications and Technology Community



Customizing Machine Chat Translation: Improvements are Quantifiably Better

TMCnews Featured Article

July 20, 2011

Customizing Machine Chat Translation: Improvements are Quantifiably Better

By David Sims, TMCnet Contributing Editor

With some articles, all you need to read is the headline to get the tone, attitude and gist of what’s going on. For example: “Is Machine Chat Translation like a 2-Year Old or a 45-Year Old?”

Industry observer Saul Marcus wrote recently how Andrew McAfee, a research scientist at MIT (News - Alert) who studies how digital technology affects the business world, asked him that very question. Marcus says McAfee believes “high quality” machine translation will have a huge impact on business.

Note the qualifier, “high quality.” As in, not the two-year old variety. “By customizing language models with the terminology and phraseology of the domain of interest, you can elevate the machine translation quailty for content (such as FAQs, knowledgebases, etc.) specific to that domain,” Marcus says, adding that such improvements are genuinely quantifiable: “With customization, you can train machine translation language models to produce higher quality, more actionable translation.”

What McAfee (News - Alert) told Marcus is that if you immerse a two-year old in a foreign language environment, “that child will be fluent in a matter of a couple of years if not months. If you immerse a forty-five year old in a foreign country, he or she might never become fluent.”

Almost one year ago TMCnet reported that AppTek, a provider of human language technology “HLT,” has released Chinese language for its hybrid machine translation “HMT” system.

According to company officials, AppTek’s supported languages are being migrated to its patent-pending Hybrid Machine Translation platform, while at the same time AppTek is continuously introducing more new languages all through the year.

In addition, AppTek's Hybrid Machine Translation solution, which is a complete combination of statistical and rule-based methodologies, has been ranked the best for the last three years in an evaluation of machine translation systems by the National Institute of Standards and Technology.

AppTek's HMT is a true advancement in the field of machine translation and customers will be able to notice a significant improvement in accuracy, abundance of information and fluency. The HMT product offers the same capabilities as offered by other rule-based or statistical MT systems in one strong platform which can run on a server or a detached laptop.

David Sims is a contributing editor for TMCnet. To read more of David’s articles, please visit his columnist page. He also blogs for TMCnet here.

Edited by Jamie Epstein

Technology Marketing Corporation

35 Nutmeg Drive Suite 340, Trumbull, Connecticut 06611 USA
Ph: 800-243-6002, 203-852-6800
Fx: 203-866-3326

General comments:
Comments about this site:


© 2018 Technology Marketing Corporation. All rights reserved | Privacy Policy