Lexical knowledge acquisition from bilingual corpora

Abstract

For practical research in natural language processing, it is indispensabh . to develop a large scale sentantic dictionary for computers. It is especially important to improve the tecbuiques ibr compiling semantic dictionaries torn natural language texts such as those in existing human dictionaries or in large corpora. 11owever } there are at least two ditIicultics in analyzing existing texts: tbe problem of syntactlc ambiguities and the problem of polysemy. Our approach to solw, these difficulties is to make use of translation exam- pies in two dis[inet languagcs that haw (luitc different sylltactic structures and word mealtings. The reason we took tbis approach is that in many cmcs both syn tactic alrd scnlantic ambiguities arc resolved by comparing analyzed results from botb lauguages. In this paper, we propose a method ibr resolving the syntac tie ambiguities of translation examples of bihngual corpora and a method for acquiring lexical knowledge, sucb a.s case frames of verbs and attribute sets of noutts.