Machine Translation

Overview

Machine translation is the application of computers to the translation of texts from one natural language into another. In its broadest sense MT can be understood to include such computer applications as compilers and compression programs etc., which convert a file i25 February, 2008. However, what we are interested in is natural language processing (NLP). Machine translation is a main research area of NLPlab.

Machine Translation Techniques:

• Interlingual MT
• Knowledge-based MT(KBMT)
• Language-based MT (LBMT)
• Example-based MT (EBMT)

Our Research Fruits

In 1980, our lab first got involved in the area of machine translation.
In 1984, we designed and implemented a Chinese-English machine translation model based on Schank’s conceptual depending theory.
From 1987 to 1994, we took part in the research of CICC Asian multilingual machine translation system.
From 1995 to 2000, we initiated the research on Chinese-English-Korean Multilingual Machine Translation system.
In 2002, we cooperated with Korean Pohang University of Science and Technology to develop “Chinese-Korean Machine Translation System.

Current Research Direction

Our current research on MT focuses on EBMT. The basic idea is to collect a bilingual corpus of translation pairs and then use a best match algorithm to find the closest example to the source phrase in question. This gives a translation template, which can then be filled in by word-for-word translation.

We’ve developed several MT systems based on EBMT. We applied new and efficient algorithms in our systems and improved the performance.

Current Project: E-Chunk-based Machine Translation