2018 Fiscal Year Final Research Report
The Extraction and Application of Visual and Semantic Information from Japanese Treebanks with Syntactic and Semantic Analysis Annotation
| Project/Area Number |
15K02469
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Linguistics
|
|---|
| Research Institution | Hirosaki University (2018)
National Institute for Japanese Language and Linguistics (2016-2017)
Tohoku University (2015) |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2015-04-01 – 2019-03-31
|
| Keywords | semantic dependencies / parsed corpus / visualisation / annotation / predicate arguments / discourse relations |
|---|
| Outline of Final Research Achievements |
The research aim has been to develop methods of visualising and making accessible semantic information from analyses of Japanese and English, notably: connective, predicate, argument, and operator-binding information.
The key result of this work has been the development of a visualisation tool for semantic relationships derivable from parsed corpus data. This tool has the capability to present many relationships found internally to sentences and across discourse, e.g., providing a way to visualise semantic roles, and anaphoric dependencies. This aids human annotators building a parsed corpus to assess whether their interpretations of sentences (and discourse) have been adequately captured by the annotation.
|
| Free Research Field |
linguistics
|
| Academic Significance and Societal Importance of the Research Achievements |
視覚化ツールは次の解析情報付きコーパスの構築で利用されている：(1) 英語（TSPC，7026ツリー，87182語，http://www.compling.jp/ajb129/tspc.html），(2) 現代日本語（NPCMJ，30460ツリー，505319語，http://npcmj.ninjal.ac.jp/interfaces), (3) 上代日本語（MYS97，159ツリー，2549語，http://www.compling.jp/mys97）。これにより，以前は不可能であった依存関係の層を視覚的に示すということが、数千の実例という規模で可能になった。
|
