Research interest

Research interest

• • Natural Language Processing

  • Machine Translation

  • Knowledge Representation

  • Interlingua (Language Intermediate Representation)

  • Corpus-based Approach

  • Stochastic Approach

  • Syntactic and Morphological Analysis

Some interesting information concerning the Thai language:

Dictionary is the fundamental knowledge base for language understanding. We put a lot of efforts in building the lexicon base for word reference in Thai language. In the process of making dictionary at present time, we cannot just sit down thinking of word, try defining its definition and writing a sentence of the usage for it. Thanks to the computer technology, it is now available to testify our supposition of the word entry and its use in the praticle text (text corpora). On the other hand, making use of text corpora allows us to be able to extract the more practical word entry and define the use without the lexicographer's individual distortion (rare words are coded rather than the common words).

• 1. LEXiTRON

     1. LINKS first released an electronic dictionary, called LEXiTRON, in 1995. The name, LEXiTRON comes from lexicon + electron (lexicon is a greek word of lexikon, means the dictionary or the set of all the words and idioms of any language; and electron means the elementary particle, implies the eletronic media of storage). The LEXiTRON Ver 1.1 is the result of our very first effort to build a dictionary that provides the semantically related information of words as well as the way of the access.

     2. Each word includes the information of :-

        • Pronunciation

        • Part-of-speech

        • Usage (a sample sentence for each sense)

          • • Verb pattern for Verb

            • Classifier for Noun

        • Synonyms-Anotonyms

        • Word group

        • English equivalent

     3. These informations are extracted from text corpora according to the result of statistical analysis of word occurrence. Though the total number of word in theLEXiTRON Ver 1.1 is very few, they are all extracted from the text corpora. The second version of LEXiTRON is now on process with the expectation that it will cover most of the Thai words (they are better called regid expressions or open compounds, because there is no rigorous definition of what a word or a sentence in the Thai language is).

     4. Contact LINKS or me to get more information about LEXiTRON. Web based service of LEXiTRON is also available.

     5. (LEXiTRON is the registered trademark of NECTEC.)

• 1. Thai Royal Institute Dictionary (TRID)

     1. TRID is one of the most popular Thai dictionaries. The first edition was published in B.E.2525 (A.D.1982). In the mid of 1995, the 5th edition of B.E.2525 was published with the purpose to correct the mistakes, to redefine some words in the categories of pronoun and exclamation, and also to add some new words.

• • • Statistical analysis of the basic term dictionary of Thai Royal Institute (TRID)

• 2. Some Statistical Data on the Thai Language

  3. ParSit

• • 1. "A quick-look translation service for cross language navigation." ParSit is developed based on NEC Crossroad machine translation system.