The noun phrase `NP `no' NP', that consists of two noun phrases connected by an adnominal particles `no', is frequently used in Japanese sentences. The surface structure of this phrase is simple, but it has various semantic structures. There has been proposed a grammar to get the semantic structure of this phrase systematically from its syntactic structure. And it is important to get a valid syntactic structure of an input noun phrase. However, there are several syntactic structures corresponding to the input noun phrase. This paper presents a Probabilistic Context Free Grammar (PCFG), which has coocurrence constraints embedded by subdividing a nonterminal with the semantic category of the headword of the phrase derived from it. Then the result of the experiment shows the effectiveness of this PCFG.
|Number of pages||6|
|Journal||Research Reports on Information Science and Electrical Engineering of Kyushu University|
|Publication status||Published - Mar 1 2000|
All Science Journal Classification (ASJC) codes
- Computer Science(all)
- Electrical and Electronic Engineering