ENGLISH-JAPANESE MACHINE TRANSLATION: VERB-BASED NOUN CLASSIFICATION APPROACH
Chai Kiat Yeo, WAI KONG LAM, Ing Yann Soon · International Journal of Artificial Intelligence Tools · 1996
A new approach to machine translation, capable of resolving different meanings of a verb in sentences of varying context, is described. The design revolves around the Verb Usage Frame (VUF) and the Noun Classification Hierarchy (NCH). VUF contains different context items which embody the different semantic usages of a verb under different contexts. The meaning of the verb is resolved through the classifications of its subject and object, achieved through the NCH. NCH returns not just the basic classification of a noun but also its super-classification. This allows thorough semantic analysis of both the verb and the noun. The entire design is implemented using object-oriented techniques and a prototype English-Japanese machine translator is built to illustrate the merits of the design.