2017 Fiscal Year Final Research Report
Natural sentence generation by syntactic processing that fails like humans
Project/Area Number |
26580082
|
Research Category |
Grant-in-Aid for Challenging Exploratory Research
|
Allocation Type | Multi-year Fund |
Research Field |
Linguistics
|
Research Institution | Shizuoka University |
Principal Investigator |
|
Project Period (FY) |
2014-04-01 – 2018-03-31
|
Keywords | 自然言語処理 / 対話システム / 心理学的妥当性 |
Outline of Final Research Achievements |
Typical natural language processing task tries to recover human annotations by some automatic tools. However, it is still difficult to generate sentences. In addition, most tools assume written language, which performance decrease when applied to spoken language. I dare limit our system model to fail like humans. Based on our psychologically plausible model, I build a syntactic parser with semantic analysis extention. I evaluated my system by applying to automatic dectection of dialog breakdowns, dialog system for AI werewolf agents, etc.
|
Free Research Field |
自然言語処理
|