| 講演抄録/キーワード |
| 講演名 |
2014-12-15 13:10
[招待講演]Dialogue state tracking in statistical dialogue management ○Kai Yu・Lu Chen(SJTU) SP2014-108 |
| 抄録 |
(和) |
(事前公開アブストラクト) Dialogue state tracking (DST) is a process to estimate the distribution of
the dialogue states of each dialogue turn given the previous interaction
history. It plays an important role in statistical dialogue management
based on partially observable Markov decision process (POMDP). This paper
reviews and compare various statistical and rule based approaches for
dialogue state tracking. Then, a novel framework, constrained Markov
Polynomial Bayesian, is introduced. It combines the advantage of both rule
and statistical approach and offers a highly efficient way for dialogue
state tracking. |
| (英) |
(Advance abstract in Japanese is available) |
| キーワード |
(和) |
/ / / / / / / |
| (英) |
/ / / / / / / |
| 文献情報 |
信学技報, vol. 114, no. 365, SP2014-108, pp. 25-29, 2014年12月. |
| 資料番号 |
SP2014-108 |
| 発行日 |
2014-12-08 (SP) |
| ISSN |
Print edition: ISSN 0913-5685 Online edition: ISSN 2432-6380 |
著作権に ついて |
技術研究報告に掲載された論文の著作権は電子情報通信学会に帰属します.(許諾番号:10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) |
| PDFダウンロード |
SP2014-108 |