Presentation | 2017-12-20 Trial of Reinforcement Learning to Analyze Command Utterances for In-vehicle Computer Masato Tokuhisa, Shuhei Kimura, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | This paper reports a trial of reinforcement learning to analyze command utterances for an in-vehicle computer. The computer gives such services as music, tourism guide, weather report, and so on. Since the users utter additional or revisal words in their utterances, the utterance analyzer needs to accept the illegal sentences. We propose recognition-actions that extract information or shift the focus to the word of the utterance so as to use reinforcement learning, which can learn the sequences of actions. The experimental results show the accuracy is 98.2% to analyze 900 utterances that cover 85 kinds of methods for the in-vehicle computer's services. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | in-vehivle computer / utterance analysis / reinforcement learning / Q-learning / action for recognition |
Paper # | NLC2017-38 |
Date of Issue | 2017-12-13 (NLC) |
Conference Information | |
Committee | NLC / IPSJ-NL / SP / IPSJ-SLP |
---|---|
Conference Date | 2017/12/20(3days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Waseda Univ. Green Computing Systems Research Organization |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | The 4th Natural Language Processing Symposium & The 19th Spoken Language Symposium |
Chair | Hiroshi Kanayama(IBM) / Kentaro Inui(Tohoku Univ.) / Yoichi Yamashita(Ritsumeikan Univ.) / Nobuaki Minematsu(Univ. Tokyo) |
Vice Chair | Takeshi Sakaki(Hottolink) / Kazutaka Shimada(Kyushu Inst. of Tech.) / / Hiroki Mori(Utsunomiya Univ.) |
Secretary | Takeshi Sakaki(Ryukoku Univ.) / Kazutaka Shimada(NTT) / (Osaka Univ.) / Hiroki Mori(Tokyo Inst. of Tech.) / (Mixi Co. Ltd.) |
Assistant | Mitsuo Yoshida(Toyohashi Univ. of Tech.) / Takeshi Kobayakawa(NICT) / / Kei Hashimoto(Nagoya Inst. of Tech.) / Satoshi Kobashikawa(NTT) |
Paper Information | |
Registration To | Technical Committee on Natural Language Understanding and Models of Communication / Special Interest Group on Natural Language / Technical Committee on Speech / Special Interest Group on Spoken Language Processing |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Trial of Reinforcement Learning to Analyze Command Utterances for In-vehicle Computer |
Sub Title (in English) | |
Keyword(1) | in-vehivle computer |
Keyword(2) | utterance analysis |
Keyword(3) | reinforcement learning |
Keyword(4) | Q-learning |
Keyword(5) | action for recognition |
1st Author's Name | Masato Tokuhisa |
1st Author's Affiliation | Tottori University(Tottori Univ.) |
2nd Author's Name | Shuhei Kimura |
2nd Author's Affiliation | Tottori University(Tottori Univ.) |
Date | 2017-12-20 |
Paper # | NLC2017-38 |
Volume (vol) | vol.117 |
Number (no) | NLC-367 |
Page | pp.pp.55-60(NLC), |
#Pages | 6 |
Date of Issue | 2017-12-13 (NLC) |