Automatic template feature extraction and the application to utterance in a dialogue system

Yoshitaka Mikami, Masafumi Hagiwara

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we propose an automatic template features extraction method and apply it to utterance generation in a dialogue system. Template-based utterance generation has been widely used in many dialogue systems because of its robustness. Although variety of templates and the appropriate selection are crucial points in the method, they have not been paid attention so far. This paper focuses on the points; first, we propose the new neural network model utilizingLSTM (Long Short-Term Memory) to extract effective and unique features for templates, and then applied it to utterance generation in a dialogue system. To examine the effectiveness of the proposed method, we conduct two kinds of experiments; subjective evaluation and dialogue breakdown detection experiment. In both of the experiments, the proposed method has shown higher accuracy than the conventional methods.

Original languageEnglish
Title of host publication2nd International Conference on Machine Learning and Soft Computing, ICMLSC 2018
PublisherAssociation for Computing Machinery
Pages164-168
Number of pages5
ISBN (Electronic)9781450363365
DOIs
Publication statusPublished - 2018 Feb 2
Event2nd International Conference on Machine Learning and Soft Computing, ICMLSC 2018 - Phu Quoc Island, Viet Nam
Duration: 2018 Feb 22018 Feb 4

Other

Other2nd International Conference on Machine Learning and Soft Computing, ICMLSC 2018
CountryViet Nam
CityPhu Quoc Island
Period18/2/218/2/4

Fingerprint

Feature extraction
Experiments
Neural networks
Long short-term memory

Keywords

  • Long short-term memory
  • Sentence embeddings
  • Template based dialogue system
  • Utterance generation

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Computer Networks and Communications
  • Computer Vision and Pattern Recognition
  • Software

Cite this

Mikami, Y., & Hagiwara, M. (2018). Automatic template feature extraction and the application to utterance in a dialogue system. In 2nd International Conference on Machine Learning and Soft Computing, ICMLSC 2018 (pp. 164-168). Association for Computing Machinery. https://doi.org/10.1145/3184066.3184069

Automatic template feature extraction and the application to utterance in a dialogue system. / Mikami, Yoshitaka; Hagiwara, Masafumi.

2nd International Conference on Machine Learning and Soft Computing, ICMLSC 2018. Association for Computing Machinery, 2018. p. 164-168.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Mikami, Y & Hagiwara, M 2018, Automatic template feature extraction and the application to utterance in a dialogue system. in 2nd International Conference on Machine Learning and Soft Computing, ICMLSC 2018. Association for Computing Machinery, pp. 164-168, 2nd International Conference on Machine Learning and Soft Computing, ICMLSC 2018, Phu Quoc Island, Viet Nam, 18/2/2. https://doi.org/10.1145/3184066.3184069
Mikami Y, Hagiwara M. Automatic template feature extraction and the application to utterance in a dialogue system. In 2nd International Conference on Machine Learning and Soft Computing, ICMLSC 2018. Association for Computing Machinery. 2018. p. 164-168 https://doi.org/10.1145/3184066.3184069
Mikami, Yoshitaka ; Hagiwara, Masafumi. / Automatic template feature extraction and the application to utterance in a dialogue system. 2nd International Conference on Machine Learning and Soft Computing, ICMLSC 2018. Association for Computing Machinery, 2018. pp. 164-168
@inproceedings{3d042bfb08014e5ca61eec07d3af833b,
title = "Automatic template feature extraction and the application to utterance in a dialogue system",
abstract = "In this paper, we propose an automatic template features extraction method and apply it to utterance generation in a dialogue system. Template-based utterance generation has been widely used in many dialogue systems because of its robustness. Although variety of templates and the appropriate selection are crucial points in the method, they have not been paid attention so far. This paper focuses on the points; first, we propose the new neural network model utilizingLSTM (Long Short-Term Memory) to extract effective and unique features for templates, and then applied it to utterance generation in a dialogue system. To examine the effectiveness of the proposed method, we conduct two kinds of experiments; subjective evaluation and dialogue breakdown detection experiment. In both of the experiments, the proposed method has shown higher accuracy than the conventional methods.",
keywords = "Long short-term memory, Sentence embeddings, Template based dialogue system, Utterance generation",
author = "Yoshitaka Mikami and Masafumi Hagiwara",
year = "2018",
month = "2",
day = "2",
doi = "10.1145/3184066.3184069",
language = "English",
pages = "164--168",
booktitle = "2nd International Conference on Machine Learning and Soft Computing, ICMLSC 2018",
publisher = "Association for Computing Machinery",

}

TY - GEN

T1 - Automatic template feature extraction and the application to utterance in a dialogue system

AU - Mikami, Yoshitaka

AU - Hagiwara, Masafumi

PY - 2018/2/2

Y1 - 2018/2/2

N2 - In this paper, we propose an automatic template features extraction method and apply it to utterance generation in a dialogue system. Template-based utterance generation has been widely used in many dialogue systems because of its robustness. Although variety of templates and the appropriate selection are crucial points in the method, they have not been paid attention so far. This paper focuses on the points; first, we propose the new neural network model utilizingLSTM (Long Short-Term Memory) to extract effective and unique features for templates, and then applied it to utterance generation in a dialogue system. To examine the effectiveness of the proposed method, we conduct two kinds of experiments; subjective evaluation and dialogue breakdown detection experiment. In both of the experiments, the proposed method has shown higher accuracy than the conventional methods.

AB - In this paper, we propose an automatic template features extraction method and apply it to utterance generation in a dialogue system. Template-based utterance generation has been widely used in many dialogue systems because of its robustness. Although variety of templates and the appropriate selection are crucial points in the method, they have not been paid attention so far. This paper focuses on the points; first, we propose the new neural network model utilizingLSTM (Long Short-Term Memory) to extract effective and unique features for templates, and then applied it to utterance generation in a dialogue system. To examine the effectiveness of the proposed method, we conduct two kinds of experiments; subjective evaluation and dialogue breakdown detection experiment. In both of the experiments, the proposed method has shown higher accuracy than the conventional methods.

KW - Long short-term memory

KW - Sentence embeddings

KW - Template based dialogue system

KW - Utterance generation

UR - http://www.scopus.com/inward/record.url?scp=85060501275&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85060501275&partnerID=8YFLogxK

U2 - 10.1145/3184066.3184069

DO - 10.1145/3184066.3184069

M3 - Conference contribution

SP - 164

EP - 168

BT - 2nd International Conference on Machine Learning and Soft Computing, ICMLSC 2018

PB - Association for Computing Machinery

ER -