Design and construction of a machine-tractable Malay-English lexicon
Quah, Chiew Kin.
Date of Issue2001
Biennial Conference of the Asian Association for Lexicography (2001)
School of Humanities and Social Sciences
In this paper, we introduce a machine-tractable Malay-English lexicon currently being developed at NTT MSC, Malaysia. The lexicon is designed to satisfy three criteria: Firstly, to develop and document detailed syntactic features useful for both analysis and generation. Secondly, to use a well-developed semantic ontology, in our case the semantic classes from the 2,710 classes used in the machine-tractable Japanese-English Goi-Taikei ontology. Thirdly, to get as wide cover as possible. The lexicon currently contains around 91,000 Malay-English pairs. Each entry consists of nine major fields, which include such information as numeral classifiers associated with common nouns, and meta-codes to show honorific use, register and origin. In addition, English and Chinese translations and comments are provided for future use in machine translation systems and also as an aid for non-Malay speakers. A version of the dictionary, which does not show all fields, is available on-line.