Analysis of Syntactic Structure of Japanese Compound Nouns

this paper, we propose a method to identify the syntactic structure of compound nouns based on the syntactic categories of constituents nouns. In our previous work, we assumed that a modification structure of a compound noun can be represented by a binary tree. We also assumed that the property of the right branch of a (sub)tree represents the property of the (sub)tree itself. This assumption is based on the fact that Japanese is a head-final language, namely, a modifier comes on the left of its modifiee. Keeping these assumptions, we can determine a syntactic structure of a compound noun by identifying the syntactic relation between each modifier and modifiee pair. In this paper, we focus on identifying the syntactic relation between two nouns (a modifier and a modifiee) . For example, "kankyo-hogo" would be analyzed that "kankyo (environment)" is a complement (in particular, object) of "hogo (conservation)". 2 Identifying syntactic relation