Extending predicate-argument thesaurus with large-scale language resources for applying to natural language processing and linguistic analysis
| Project/Area Number |
26370485
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Linguistics
|
|---|
| Research Institution | Okayama University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2014-04-01 – 2017-03-31
|
| Project Status |
Completed (Fiscal Year 2016)
|
| Budget Amount *help |
¥3,120,000 (Direct Cost: ¥2,400,000、Indirect Cost: ¥720,000)
Fiscal Year 2016: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2015: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2014: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
|
|---|
| Keywords | 項構造 / 名詞項構造 / 述語項構造 / 意味役割 / 述語項構造シソーラス / PropBank / コピュラ / Generative Lexicon / 名詞述語文 / 類義語データベース / 述語シソーラス |
|---|
| Outline of Final Research Achievements |
We developed and published a new framework of nominal argument structure for Japanese. The key technique of describing arguments is number-based semantic roles, which enables us to identify each argument between paraphrased sentences in nominal predicates. We constructed 5,000 example sentences of nominal predicates; those of 13000 sentences have semantic role labels and links to the corresponding examples in the predicate thesaurus. We also constructed paraphrase data between different part-of-speech words and incorporate the paraphrase data into an argument structure analyzer. Besides, we improve the performance of the argument structure analyzer by applying a statistical learning method, and published the results in IPSJ Journal and the proceedings of International Conference PACLING 2015. We also show that the argument structure analyzer with nominal predicates enables us to improve performance of Japanese textual entailment task.
|
Report
(4 results)
Research Products
(15 results)
