| Presentation | 2018-12-12 Data augmentation using stereotypical reply for patients' tweet identification Reine Asakawa, Tomoyosi Akiba, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In this study, we try to identify patients' tweets for symptom surveillance using Twitter. This functionality is indispensable for developing a system Identifying disease epidemic. Most previous work employed a supervised machine learning methods. In general, they need a large amount of labeled corpus, which are very expensive to be created. In order to cope with this problem, we proposed a method to automatically acquire training corpus from Twitter by using a typical response to a patient. In this paper, we propose a data augmentation approach that extends a training data for RNN-based patient identifier with those automatically acquired corpus. The method consists of two steps. As the first step, initial parameters of identifier are trained by the automatically required large corpus. As the Second step, they are continuously trained by using a small amount of training corpus annotated manually. By this method, it is possible to effectively combine two kinds of corpus in a manner complementing each other. We experimented to apply the proposed data augmentation method for the training of RNN-based patient identifiers. The result showed the proposed model successfully improved the identification performance over the model without data augmentation. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | RNN / Twitter / DataAugmentation / Fine-tuning |
| Paper # | NLC2018-31 |
| Date of Issue | 2018-12-04 (NLC) |
| Conference Information | |
| Committee | NLC / IPSJ-NL / SP / IPSJ-SLP |
|---|---|
| Conference Date | 2018/12/10(3days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Waseda Univ. Nishiwaseda Campus |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | The 5th Natural Language Processing Symposium & The 20th Spoken Language Symposium |
| Chair | Takeshi Sakaki(Hottolink) / / Yoichi Yamashita(Ritsumeikan Univ.) |
| Vice Chair | Mitsuo Yoshida(Toyohashi Univ. of Tech.) / Kazutaka Shimada(Kyushu Inst. of Tech.) / / Akinobu Ri(Nagoya Inst. of Tech.) |
| Secretary | Mitsuo Yoshida(Ryukoku Univ.) / Kazutaka Shimada(NTT) / / Akinobu Ri(Kyoto Univ.) / (Meijo Univ.) |
| Assistant | Takeshi Kobayakawa(NHK) / Hiroki Sakaji(Univ. of Tokyo) / / Tomoki Koriyama(Tokyo 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) | Data augmentation using stereotypical reply for patients' tweet identification |
| Sub Title (in English) | |
| Keyword(1) | RNN |
| Keyword(2) | |
| Keyword(3) | DataAugmentation |
| Keyword(4) | Fine-tuning |
| 1st Author's Name | Reine Asakawa |
| 1st Author's Affiliation | Toyohashi University of Technology(TUT) |
| 2nd Author's Name | Tomoyosi Akiba |
| 2nd Author's Affiliation | Toyohashi University of Technology(TUT) |
| Date | 2018-12-12 |
| Paper # | NLC2018-31 |
| Volume (vol) | vol.118 |
| Number (no) | NLC-355 |
| Page | pp.pp.55-60(NLC), |
| #Pages | 6 |
| Date of Issue | 2018-12-04 (NLC) |