I annotated semantic information on copular sentences contained in the Kyoto University Text Corpus for the purpose of consolidating descriptive research and constructing a sharable resource for linguistic research. This task had four subtasks: i) converting the corpus into XML format, ii) annotating word meanings using four semantic resources (Extended Named Entity Annotated Corpora, CRL named entity data, Japanese WordNet, and SUMO), iii) extracting copular sentences in the corpus, and iv) annotating a semantic relation between the subject and predicate of copular sentences. Following the results of the annotation, I investigated the semantic relationships between the subject and predicate and their word meanings, and the syntactic and semantic features of Japanese copular sentences.
日本語名詞述語文への意味情報付与
papers0804.pdf
(936.9KB)
概要
