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

Publication series

NameACM International Conference Proceeding Series

Other

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

Keywords

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

ASJC Scopus subject areas

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

Fingerprint

Dive into the research topics of 'Automatic template feature extraction and the application to utterance in a dialogue system'. Together they form a unique fingerprint.

Cite this