Speech act identification using an ontology-based partial pattern tree

Chung Hsien Wu, Jui Feng Yeh, Ming Jun Chen

研究成果: Paper

1 引文 (Scopus)

摘要

This paper presents an ontology-based partial pattern tree to identify the speech act in a spoken dialogue system. This study first extracts the key words/concepts in an application domain using latent semantic analysis (LSA). A partial pattern tree is used to deal with the ill-formed sentence problem in a spoken dialogue system. Concept expansion based on domain ontology is adopted to improve system performance. For performance evaluation, a medical dialogue system with multiple services, including registration information, clinic information and FAQ information, is implemented. Four performance measures were separately used for evaluation. The speech act identification rate achieves 86.2%. A Task Success Rate of 77% is obtained. The contextual appropriateness of the system response is 78.5%. Finally, the correct rate for FAQ retrieval is 82% with an improvement of 15% in comparison with the keyword-based vector space model. The results show the proposed ontology-based partial pattern tree is effective for dialogue management.

原文English
頁面2157-2160
頁數4
出版狀態Published - 2004 一月 1
事件8th International Conference on Spoken Language Processing, ICSLP 2004 - Jeju, Jeju Island, Korea, Republic of
持續時間: 2004 十月 42004 十月 8

Other

Other8th International Conference on Spoken Language Processing, ICSLP 2004
國家Korea, Republic of
城市Jeju, Jeju Island
期間04-10-0404-10-08

指紋

speech act
ontology
communication technology
performance
evaluation
dialogue
semantics
management
Speech Acts
Ontology
Spoken Dialogue Systems
Key Words

All Science Journal Classification (ASJC) codes

  • Language and Linguistics
  • Linguistics and Language

引用此文

Wu, C. H., Yeh, J. F., & Chen, M. J. (2004). Speech act identification using an ontology-based partial pattern tree. 2157-2160. 論文發表於 8th International Conference on Spoken Language Processing, ICSLP 2004, Jeju, Jeju Island, Korea, Republic of.
Wu, Chung Hsien ; Yeh, Jui Feng ; Chen, Ming Jun. / Speech act identification using an ontology-based partial pattern tree. 論文發表於 8th International Conference on Spoken Language Processing, ICSLP 2004, Jeju, Jeju Island, Korea, Republic of.4 p.
@conference{e09cc6e30a2542ed960a83ed8a14ed78,
title = "Speech act identification using an ontology-based partial pattern tree",
abstract = "This paper presents an ontology-based partial pattern tree to identify the speech act in a spoken dialogue system. This study first extracts the key words/concepts in an application domain using latent semantic analysis (LSA). A partial pattern tree is used to deal with the ill-formed sentence problem in a spoken dialogue system. Concept expansion based on domain ontology is adopted to improve system performance. For performance evaluation, a medical dialogue system with multiple services, including registration information, clinic information and FAQ information, is implemented. Four performance measures were separately used for evaluation. The speech act identification rate achieves 86.2{\%}. A Task Success Rate of 77{\%} is obtained. The contextual appropriateness of the system response is 78.5{\%}. Finally, the correct rate for FAQ retrieval is 82{\%} with an improvement of 15{\%} in comparison with the keyword-based vector space model. The results show the proposed ontology-based partial pattern tree is effective for dialogue management.",
author = "Wu, {Chung Hsien} and Yeh, {Jui Feng} and Chen, {Ming Jun}",
year = "2004",
month = "1",
day = "1",
language = "English",
pages = "2157--2160",
note = "8th International Conference on Spoken Language Processing, ICSLP 2004 ; Conference date: 04-10-2004 Through 08-10-2004",

}

Wu, CH, Yeh, JF & Chen, MJ 2004, 'Speech act identification using an ontology-based partial pattern tree', 論文發表於 8th International Conference on Spoken Language Processing, ICSLP 2004, Jeju, Jeju Island, Korea, Republic of, 04-10-04 - 04-10-08 頁 2157-2160.

Speech act identification using an ontology-based partial pattern tree. / Wu, Chung Hsien; Yeh, Jui Feng; Chen, Ming Jun.

2004. 2157-2160 論文發表於 8th International Conference on Spoken Language Processing, ICSLP 2004, Jeju, Jeju Island, Korea, Republic of.

研究成果: Paper

TY - CONF

T1 - Speech act identification using an ontology-based partial pattern tree

AU - Wu, Chung Hsien

AU - Yeh, Jui Feng

AU - Chen, Ming Jun

PY - 2004/1/1

Y1 - 2004/1/1

N2 - This paper presents an ontology-based partial pattern tree to identify the speech act in a spoken dialogue system. This study first extracts the key words/concepts in an application domain using latent semantic analysis (LSA). A partial pattern tree is used to deal with the ill-formed sentence problem in a spoken dialogue system. Concept expansion based on domain ontology is adopted to improve system performance. For performance evaluation, a medical dialogue system with multiple services, including registration information, clinic information and FAQ information, is implemented. Four performance measures were separately used for evaluation. The speech act identification rate achieves 86.2%. A Task Success Rate of 77% is obtained. The contextual appropriateness of the system response is 78.5%. Finally, the correct rate for FAQ retrieval is 82% with an improvement of 15% in comparison with the keyword-based vector space model. The results show the proposed ontology-based partial pattern tree is effective for dialogue management.

AB - This paper presents an ontology-based partial pattern tree to identify the speech act in a spoken dialogue system. This study first extracts the key words/concepts in an application domain using latent semantic analysis (LSA). A partial pattern tree is used to deal with the ill-formed sentence problem in a spoken dialogue system. Concept expansion based on domain ontology is adopted to improve system performance. For performance evaluation, a medical dialogue system with multiple services, including registration information, clinic information and FAQ information, is implemented. Four performance measures were separately used for evaluation. The speech act identification rate achieves 86.2%. A Task Success Rate of 77% is obtained. The contextual appropriateness of the system response is 78.5%. Finally, the correct rate for FAQ retrieval is 82% with an improvement of 15% in comparison with the keyword-based vector space model. The results show the proposed ontology-based partial pattern tree is effective for dialogue management.

UR - http://www.scopus.com/inward/record.url?scp=85009083823&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85009083823&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:85009083823

SP - 2157

EP - 2160

ER -

Wu CH, Yeh JF, Chen MJ. Speech act identification using an ontology-based partial pattern tree. 2004. 論文發表於 8th International Conference on Spoken Language Processing, ICSLP 2004, Jeju, Jeju Island, Korea, Republic of.