Latent topic modeling of word co-occurrence information for spoken document retrieval

Berlin Chen

doi:10.1109/ICASSP.2009.4960495

Latent topic modeling of word co-occurrence information for spoken document retrieval

Berlin Chen^*

^*此作品的通信作者

資訊工程學系

研究成果: 書貢獻/報告類型 › 會議論文篇章

26 引文斯高帕斯（Scopus）

摘要

In this paper, we present a word topic model (WTM) approach, discovering the co-occurrence relationship between words as well as the long-span latent topic information, for spoken document retrieval (SDR). A given document as a whole is modeled as a composite WTM model for generating an observed query. The underlying characteristics and different kinds of model structures are extensively investigated, while the performance of WTM is thoroughly analyzed and verified by comparison with a few existing retrieval models on the TDT-2 SDR task. We also attempt to incorporate part-of-speech (POS) weighting into the representations of the query observations and the WTM models for obtaining better retrieval performance.

原文	英語
主出版物標題	2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009
頁面	3961-3964
頁數	4
DOIs	https://doi.org/10.1109/ICASSP.2009.4960495
出版狀態	已發佈 - 2009
事件	2009 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2009 - Taipei, 臺灣持續時間: 2009 4月 19 → 2009 4月 24

出版系列

名字	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
ISSN（列印）	1520-6149

其他

其他	2009 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2009
國家/地區	臺灣
城市	Taipei
期間	2009/04/19 → 2009/04/24

ASJC Scopus subject areas

軟體
訊號處理
電氣與電子工程

存取文件

10.1109/ICASSP.2009.4960495

其它文件與鏈接

引用此

Chen, B. (2009). Latent topic modeling of word co-occurrence information for spoken document retrieval. 於 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009 (頁 3961-3964). 文章 4960495 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). https://doi.org/10.1109/ICASSP.2009.4960495

Latent topic modeling of word co-occurrence information for spoken document retrieval. / Chen, Berlin.
2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009. 2009. p. 3961-3964 4960495 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

研究成果: 書貢獻/報告類型 › 會議論文篇章

Chen, B 2009, Latent topic modeling of word co-occurrence information for spoken document retrieval. 於 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009., 4960495, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, 頁 3961-3964, 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2009, Taipei, 臺灣, 2009/04/19. https://doi.org/10.1109/ICASSP.2009.4960495

@inproceedings{e75ba05e583f409794cca1ea63e71549,

title = "Latent topic modeling of word co-occurrence information for spoken document retrieval",

abstract = "In this paper, we present a word topic model (WTM) approach, discovering the co-occurrence relationship between words as well as the long-span latent topic information, for spoken document retrieval (SDR). A given document as a whole is modeled as a composite WTM model for generating an observed query. The underlying characteristics and different kinds of model structures are extensively investigated, while the performance of WTM is thoroughly analyzed and verified by comparison with a few existing retrieval models on the TDT-2 SDR task. We also attempt to incorporate part-of-speech (POS) weighting into the representations of the query observations and the WTM models for obtaining better retrieval performance.",

keywords = "Language model, Probabilistic latent semantic analysis, Spoken document retrieval, Word topic model",

author = "Berlin Chen",

year = "2009",

doi = "10.1109/ICASSP.2009.4960495",

language = "English",

isbn = "9781424423545",

series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

pages = "3961--3964",

booktitle = "2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009",

note = "2009 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2009 ; Conference date: 19-04-2009 Through 24-04-2009",

}

TY - GEN

T1 - Latent topic modeling of word co-occurrence information for spoken document retrieval

AU - Chen, Berlin

PY - 2009

Y1 - 2009

N2 - In this paper, we present a word topic model (WTM) approach, discovering the co-occurrence relationship between words as well as the long-span latent topic information, for spoken document retrieval (SDR). A given document as a whole is modeled as a composite WTM model for generating an observed query. The underlying characteristics and different kinds of model structures are extensively investigated, while the performance of WTM is thoroughly analyzed and verified by comparison with a few existing retrieval models on the TDT-2 SDR task. We also attempt to incorporate part-of-speech (POS) weighting into the representations of the query observations and the WTM models for obtaining better retrieval performance.

AB - In this paper, we present a word topic model (WTM) approach, discovering the co-occurrence relationship between words as well as the long-span latent topic information, for spoken document retrieval (SDR). A given document as a whole is modeled as a composite WTM model for generating an observed query. The underlying characteristics and different kinds of model structures are extensively investigated, while the performance of WTM is thoroughly analyzed and verified by comparison with a few existing retrieval models on the TDT-2 SDR task. We also attempt to incorporate part-of-speech (POS) weighting into the representations of the query observations and the WTM models for obtaining better retrieval performance.

KW - Language model

KW - Probabilistic latent semantic analysis

KW - Spoken document retrieval

KW - Word topic model

UR - http://www.scopus.com/inward/record.url?scp=70349223893&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70349223893&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2009.4960495

DO - 10.1109/ICASSP.2009.4960495

M3 - Conference contribution

AN - SCOPUS:70349223893

SN - 9781424423545

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 3961

EP - 3964

BT - 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009

T2 - 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2009

Y2 - 19 April 2009 through 24 April 2009

ER -

Latent topic modeling of word co-occurrence information for spoken document retrieval

摘要

出版系列

其他

ASJC Scopus subject areas

存取文件

其它文件與鏈接

指紋

引用此