Supervised intensive topic models for emotion detection over short text

Yanghui RAO; Jianhui PANG; Haoran XIE; An LIU; Tak-Lam WONG; Qing LI; Fu Lee WANG

doi:10.1007/978-3-319-55753-3_26

Supervised intensive topic models for emotion detection over short text

Yanghui RAO^*
, Jianhui PANG
, Haoran XIE
, An LIU
, Tak-Lam WONG
, Qing LI
, Fu Lee WANG

^*Corresponding author for this work

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Research

7 Citations (Scopus)

Abstract

With the emergence of social media services, documents that only include a few words are becoming increasingly prevalent. More and more users post short messages to express their feelings and emotions through Twitter, Flickr, YouTube and other apps. However, the sparsity of word co-occurrence patterns in short text brings new challenges to emotion detection tasks. In this paper, we propose two supervised intensive topic models to associate latent topics with emotional labels. The first model constrains topics to relevant emotions, and then generates document-topic probability distributions. The second model establishes association among biterms and emotions by topics, and then estimates word-emotion probabilities. Experiments on short text emotion detection validate the effectiveness of the proposed models.

Original language	English
Title of host publication	Database Systems for Advanced Applications : 22nd International Conference, DASFAA 2017, Suzhou, China, March 27-30, 2017, proceedings, part I
Editors	Selçuk CANDAN, Lei CHEN, Torben Bach PEDERSEN, Lijun CHANG, Wen HUA
Publisher	Springer International Publishing AG
Pages	408-422
Number of pages	15
ISBN (Electronic)	9783319557533
ISBN (Print)	9783319557526
DOIs	https://doi.org/10.1007/978-3-319-55753-3_26
Publication status	Published - 2017
Externally published	Yes
Event	22nd International Conference on Database Systems for Advanced Applications - Suzhou, China Duration: 27 Mar 2017 → 30 Mar 2017

Publication series

Name	Lecture Notes in Computer Science
Volume	10177
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	22nd International Conference on Database Systems for Advanced Applications
Abbreviated title	DASFAA 2017
Country/Territory	China
Period	27/03/17 → 30/03/17

Bibliographical note

Publisher Copyright:
© Springer International Publishing AG 2017.

Funding

We are grateful to the anonymous reviewers for their valuable comments on this manuscript. The research has been supported by the National Natural Science Foundation of China (61502545, 61572336), two grants from the Research Grants Council of the Hong Kong Special Administrative Region, China (UGC/FDS11/E03/16 and UGC/FDS11/E06/14), the Start-Up Research Grant (RG 37/2016-2017R), and the Internal Research Grant (RG 66/2016-2017) of The Education University of Hong Kong.

Keywords

Topic model
Emotion detection
Short text analysis

Access to Document

10.1007/978-3-319-55753-3_26

Cite this

RAO, Y., PANG, J., XIE, H., LIU, A., WONG, T.-L., LI, Q., & WANG, F. L. (2017). Supervised intensive topic models for emotion detection over short text. In S. CANDAN, L. CHEN, T. B. PEDERSEN, L. CHANG, & W. HUA (Eds.), Database Systems for Advanced Applications : 22nd International Conference, DASFAA 2017, Suzhou, China, March 27-30, 2017, proceedings, part I (pp. 408-422). (Lecture Notes in Computer Science; Vol. 10177). Springer International Publishing AG. https://doi.org/10.1007/978-3-319-55753-3_26

RAO, Yanghui ; PANG, Jianhui ; XIE, Haoran et al. / Supervised intensive topic models for emotion detection over short text. Database Systems for Advanced Applications : 22nd International Conference, DASFAA 2017, Suzhou, China, March 27-30, 2017, proceedings, part I. editor / Selçuk CANDAN ; Lei CHEN ; Torben Bach PEDERSEN ; Lijun CHANG ; Wen HUA. Springer International Publishing AG, 2017. pp. 408-422 (Lecture Notes in Computer Science).

@inproceedings{7cc10c08284a4423a27b941fbb163a29,

title = "Supervised intensive topic models for emotion detection over short text",

abstract = "With the emergence of social media services, documents that only include a few words are becoming increasingly prevalent. More and more users post short messages to express their feelings and emotions through Twitter, Flickr, YouTube and other apps. However, the sparsity of word co-occurrence patterns in short text brings new challenges to emotion detection tasks. In this paper, we propose two supervised intensive topic models to associate latent topics with emotional labels. The first model constrains topics to relevant emotions, and then generates document-topic probability distributions. The second model establishes association among biterms and emotions by topics, and then estimates word-emotion probabilities. Experiments on short text emotion detection validate the effectiveness of the proposed models.",

keywords = "Topic model, Emotion detection, Short text analysis",

author = "Yanghui RAO and Jianhui PANG and Haoran XIE and An LIU and Tak-Lam WONG and Qing LI and WANG, \{Fu Lee\}",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing AG 2017.; 22nd International Conference on Database Systems for Advanced Applications, DASFAA 2017 ; Conference date: 27-03-2017 Through 30-03-2017",

year = "2017",

doi = "10.1007/978-3-319-55753-3\_26",

language = "English",

isbn = "9783319557526",

series = "Lecture Notes in Computer Science",

publisher = "Springer International Publishing AG",

pages = "408--422",

editor = "Sel{\c c}uk CANDAN and Lei CHEN and PEDERSEN, \{Torben Bach\} and Lijun CHANG and Wen HUA",

booktitle = "Database Systems for Advanced Applications : 22nd International Conference, DASFAA 2017, Suzhou, China, March 27-30, 2017, proceedings, part I",

address = "Switzerland",

}

RAO, Y, PANG, J, XIE, H, LIU, A, WONG, T-L, LI, Q & WANG, FL 2017, Supervised intensive topic models for emotion detection over short text. in S CANDAN, L CHEN, TB PEDERSEN, L CHANG & W HUA (eds), Database Systems for Advanced Applications : 22nd International Conference, DASFAA 2017, Suzhou, China, March 27-30, 2017, proceedings, part I. Lecture Notes in Computer Science, vol. 10177, Springer International Publishing AG, pp. 408-422, 22nd International Conference on Database Systems for Advanced Applications, China, 27/03/17. https://doi.org/10.1007/978-3-319-55753-3_26

Supervised intensive topic models for emotion detection over short text. / RAO, Yanghui; PANG, Jianhui; XIE, Haoran et al.
Database Systems for Advanced Applications : 22nd International Conference, DASFAA 2017, Suzhou, China, March 27-30, 2017, proceedings, part I. ed. / Selçuk CANDAN; Lei CHEN; Torben Bach PEDERSEN; Lijun CHANG; Wen HUA. Springer International Publishing AG, 2017. p. 408-422 (Lecture Notes in Computer Science; Vol. 10177).

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Research

TY - GEN

T1 - Supervised intensive topic models for emotion detection over short text

AU - RAO, Yanghui

AU - PANG, Jianhui

AU - XIE, Haoran

AU - LIU, An

AU - WONG, Tak-Lam

AU - LI, Qing

AU - WANG, Fu Lee

N1 - Publisher Copyright: © Springer International Publishing AG 2017.

PY - 2017

Y1 - 2017

N2 - With the emergence of social media services, documents that only include a few words are becoming increasingly prevalent. More and more users post short messages to express their feelings and emotions through Twitter, Flickr, YouTube and other apps. However, the sparsity of word co-occurrence patterns in short text brings new challenges to emotion detection tasks. In this paper, we propose two supervised intensive topic models to associate latent topics with emotional labels. The first model constrains topics to relevant emotions, and then generates document-topic probability distributions. The second model establishes association among biterms and emotions by topics, and then estimates word-emotion probabilities. Experiments on short text emotion detection validate the effectiveness of the proposed models.

AB - With the emergence of social media services, documents that only include a few words are becoming increasingly prevalent. More and more users post short messages to express their feelings and emotions through Twitter, Flickr, YouTube and other apps. However, the sparsity of word co-occurrence patterns in short text brings new challenges to emotion detection tasks. In this paper, we propose two supervised intensive topic models to associate latent topics with emotional labels. The first model constrains topics to relevant emotions, and then generates document-topic probability distributions. The second model establishes association among biterms and emotions by topics, and then estimates word-emotion probabilities. Experiments on short text emotion detection validate the effectiveness of the proposed models.

KW - Topic model

KW - Emotion detection

KW - Short text analysis

UR - https://www.scopus.com/pages/publications/85032290393

U2 - 10.1007/978-3-319-55753-3_26

DO - 10.1007/978-3-319-55753-3_26

M3 - Conference paper (refereed)

SN - 9783319557526

T3 - Lecture Notes in Computer Science

SP - 408

EP - 422

BT - Database Systems for Advanced Applications : 22nd International Conference, DASFAA 2017, Suzhou, China, March 27-30, 2017, proceedings, part I

A2 - CANDAN, Selçuk

A2 - CHEN, Lei

A2 - PEDERSEN, Torben Bach

A2 - CHANG, Lijun

A2 - HUA, Wen

PB - Springer International Publishing AG

T2 - 22nd International Conference on Database Systems for Advanced Applications

Y2 - 27 March 2017 through 30 March 2017

ER -

RAO Y, PANG J, XIE H, LIU A, WONG TL, LI Q et al. Supervised intensive topic models for emotion detection over short text. In CANDAN S, CHEN L, PEDERSEN TB, CHANG L, HUA W, editors, Database Systems for Advanced Applications : 22nd International Conference, DASFAA 2017, Suzhou, China, March 27-30, 2017, proceedings, part I. Springer International Publishing AG. 2017. p. 408-422. (Lecture Notes in Computer Science). Epub 2017 Mar 22. doi: 10.1007/978-3-319-55753-3_26

Supervised intensive topic models for emotion detection over short text

Abstract

Publication series

Conference

Bibliographical note

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Fast supervised topic models for short text emotion detection

Cite this