Multilingual generation of numeral classifiers using a common ontology

要旨

In this paper, we present a solution to the problem of generating both Japanese and Korean
\n
\n
\n
\n
\n
\n
\n
\nnumeral classifiers using semantic classes from an ontology. Most nouns must use a numeral
\n
\n
\n
\n
\n
\n
\n
\nclassifier when they are quantified in languages such as Chinese, Japanese, Korean, Malay and
\n
\n
\n
\n
\n
\n
\n
\nThai. In order to select an appropriate classifier, we propose an algorithm which associates
\n
\n
\n
\n
\n
\n
\n
\nclassifiers with semantic classes and uses inheritance to list only exceptional classifiers with
\n
\n
\n
\n
\n
\n
\n
\nindividual nouns. The algorithm generates sortal classifiers with an accuracy of 81%. We reuse
\n
\n
\n
\n
\n
\n
\n
\nthe ontology provided by Goi-Taikei — a Japanese lexicon, and show that it is a reasonable
\n
\n
\n
\n
\n
\n
\n
\nchoice for this task, requiring information to be entered for less than 6% of individual nouns.

主題