Build emotion lexicon from the mood of crowd via topic-assisted joint non-negative matrix factorization

Kaisong Song, Wei Gao, Ling Chen, Shi Feng, Daling Wang, Chengqi Zhang

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

8 Citations (Scopus)

Abstract

In the research of building emotion lexicons, we witness the exploitation of crowd-sourced affective annotation given by readers of online news articles. Such approach ignores the relationship between topics and emotion expressions which are often closely correlated. We build an emotion lexicon by developing a novel joint non-negative matrix factorization model which not only incorporates crowd-annotated emotion labels of articles but also generates the lexicon using the topic-specific matrices obtained from the factorization process. We evaluate our lexicon via emotion classification on both benchmark and built-in-house datasets. Results demonstrate the high-quality of our lexicon.

Original languageEnglish
Title of host publicationSIGIR 2016 - Proceedings of the 39th International ACM SIGIR Conference on Research and Development in Information Retrieval
PublisherAssociation for Computing Machinery, Inc
Pages773-776
Number of pages4
ISBN (Electronic)9781450342902
DOIs
Publication statusPublished - 7 Jul 2016
Event39th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2016 - Pisa, Italy
Duration: 17 Jul 201621 Jul 2016

Other

Other39th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2016
CountryItaly
CityPisa
Period17/7/1621/7/16

Fingerprint

Factorization
Labels

Keywords

  • Emotion classification
  • Emotion lexicon
  • Joint NMF

ASJC Scopus subject areas

  • Information Systems
  • Software

Cite this

Song, K., Gao, W., Chen, L., Feng, S., Wang, D., & Zhang, C. (2016). Build emotion lexicon from the mood of crowd via topic-assisted joint non-negative matrix factorization. In SIGIR 2016 - Proceedings of the 39th International ACM SIGIR Conference on Research and Development in Information Retrieval (pp. 773-776). Association for Computing Machinery, Inc. https://doi.org/10.1145/2911451.2914759

Build emotion lexicon from the mood of crowd via topic-assisted joint non-negative matrix factorization. / Song, Kaisong; Gao, Wei; Chen, Ling; Feng, Shi; Wang, Daling; Zhang, Chengqi.

SIGIR 2016 - Proceedings of the 39th International ACM SIGIR Conference on Research and Development in Information Retrieval. Association for Computing Machinery, Inc, 2016. p. 773-776.

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

Song, K, Gao, W, Chen, L, Feng, S, Wang, D & Zhang, C 2016, Build emotion lexicon from the mood of crowd via topic-assisted joint non-negative matrix factorization. in SIGIR 2016 - Proceedings of the 39th International ACM SIGIR Conference on Research and Development in Information Retrieval. Association for Computing Machinery, Inc, pp. 773-776, 39th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2016, Pisa, Italy, 17/7/16. https://doi.org/10.1145/2911451.2914759
Song K, Gao W, Chen L, Feng S, Wang D, Zhang C. Build emotion lexicon from the mood of crowd via topic-assisted joint non-negative matrix factorization. In SIGIR 2016 - Proceedings of the 39th International ACM SIGIR Conference on Research and Development in Information Retrieval. Association for Computing Machinery, Inc. 2016. p. 773-776 https://doi.org/10.1145/2911451.2914759
Song, Kaisong ; Gao, Wei ; Chen, Ling ; Feng, Shi ; Wang, Daling ; Zhang, Chengqi. / Build emotion lexicon from the mood of crowd via topic-assisted joint non-negative matrix factorization. SIGIR 2016 - Proceedings of the 39th International ACM SIGIR Conference on Research and Development in Information Retrieval. Association for Computing Machinery, Inc, 2016. pp. 773-776
@inproceedings{acc480a67610483ca6cdacdf53fa59de,
title = "Build emotion lexicon from the mood of crowd via topic-assisted joint non-negative matrix factorization",
abstract = "In the research of building emotion lexicons, we witness the exploitation of crowd-sourced affective annotation given by readers of online news articles. Such approach ignores the relationship between topics and emotion expressions which are often closely correlated. We build an emotion lexicon by developing a novel joint non-negative matrix factorization model which not only incorporates crowd-annotated emotion labels of articles but also generates the lexicon using the topic-specific matrices obtained from the factorization process. We evaluate our lexicon via emotion classification on both benchmark and built-in-house datasets. Results demonstrate the high-quality of our lexicon.",
keywords = "Emotion classification, Emotion lexicon, Joint NMF",
author = "Kaisong Song and Wei Gao and Ling Chen and Shi Feng and Daling Wang and Chengqi Zhang",
year = "2016",
month = "7",
day = "7",
doi = "10.1145/2911451.2914759",
language = "English",
pages = "773--776",
booktitle = "SIGIR 2016 - Proceedings of the 39th International ACM SIGIR Conference on Research and Development in Information Retrieval",
publisher = "Association for Computing Machinery, Inc",

}

TY - GEN

T1 - Build emotion lexicon from the mood of crowd via topic-assisted joint non-negative matrix factorization

AU - Song, Kaisong

AU - Gao, Wei

AU - Chen, Ling

AU - Feng, Shi

AU - Wang, Daling

AU - Zhang, Chengqi

PY - 2016/7/7

Y1 - 2016/7/7

N2 - In the research of building emotion lexicons, we witness the exploitation of crowd-sourced affective annotation given by readers of online news articles. Such approach ignores the relationship between topics and emotion expressions which are often closely correlated. We build an emotion lexicon by developing a novel joint non-negative matrix factorization model which not only incorporates crowd-annotated emotion labels of articles but also generates the lexicon using the topic-specific matrices obtained from the factorization process. We evaluate our lexicon via emotion classification on both benchmark and built-in-house datasets. Results demonstrate the high-quality of our lexicon.

AB - In the research of building emotion lexicons, we witness the exploitation of crowd-sourced affective annotation given by readers of online news articles. Such approach ignores the relationship between topics and emotion expressions which are often closely correlated. We build an emotion lexicon by developing a novel joint non-negative matrix factorization model which not only incorporates crowd-annotated emotion labels of articles but also generates the lexicon using the topic-specific matrices obtained from the factorization process. We evaluate our lexicon via emotion classification on both benchmark and built-in-house datasets. Results demonstrate the high-quality of our lexicon.

KW - Emotion classification

KW - Emotion lexicon

KW - Joint NMF

UR - http://www.scopus.com/inward/record.url?scp=84980329194&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84980329194&partnerID=8YFLogxK

U2 - 10.1145/2911451.2914759

DO - 10.1145/2911451.2914759

M3 - Conference contribution

SP - 773

EP - 776

BT - SIGIR 2016 - Proceedings of the 39th International ACM SIGIR Conference on Research and Development in Information Retrieval

PB - Association for Computing Machinery, Inc

ER -