Reader emotion prediction using concept and concept sequence features in news headlines

Yuanlin Yao; Ruifeng Xu; Qin Lu; Bin Liu; Jun Xu; Chengtian Zou; Li Yuan; Shuwei Wang; Lin Yao; Zhenyu He

doi:10.1007/978-3-642-54903-8_7

Reader emotion prediction using concept and concept sequence features in news headlines

Yuanlin Yao, Ruifeng Xu^*, Qin Lu, Bin Liu, Jun Xu, Chengtian Zou, Li Yuan, Shuwei Wang, Lin Yao, Zhenyu He

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

6 Citations (Scopus)

Abstract

This paper presents a method to predicate news reader emotions. News headlines supply core information of articles, thus they can serve as key information for reader emotion predication. However, headlines are always short which leads to obvious data sparseness if only lexical forms are used. To address this problem, words in their lexical forms in a headline are transferred to their concepts and concept sequence features of words in headlines based on a semantic knowledge base, namely HowNet for Chinese. These features are expected to represent the major elements which can evoke reader's emotional reactions. These transferred concepts are used with lexical features in headlines for predicating the reader's emotion. Evaluations on dataset of Sina Social News with user emotion votes show that the proposed approach which do not use any news content, achieves a comparable performance to Bag-Of-Word model using both the headlines and the news contents, making our method more efficient in reader emotion prediction.

Original language	English
Title of host publication	Computational Linguistics and Intelligent Text Processing - 15th International Conference, CICLing 2014, Proceedings
Publisher	Springer Verlag
Pages	73-84
Number of pages	12
Edition	PART 2
ISBN (Print)	9783642549021
DOIs	https://doi.org/10.1007/978-3-642-54903-8_7
Publication status	Published - 2014
Externally published	Yes
Event	15th International Conference on Computational Linguistics and Intelligent Text Processing, CICLing 2014 - Kathmandu, Nepal Duration: 6 Apr 2014 → 12 Apr 2014

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Number	PART 2
Volume	8404 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	15th International Conference on Computational Linguistics and Intelligent Text Processing, CICLing 2014
Country/Territory	Nepal
City	Kathmandu
Period	6/04/14 → 12/04/14

Keywords

Concept Feature
Concept Sequence Feature
Emotion Prediction
HowNet

Access to Document

10.1007/978-3-642-54903-8_7

Cite this

Yao, Y., Xu, R., Lu, Q., Liu, B., Xu, J., Zou, C., Yuan, L., Wang, S., Yao, L., & He, Z. (2014). Reader emotion prediction using concept and concept sequence features in news headlines. In Computational Linguistics and Intelligent Text Processing - 15th International Conference, CICLing 2014, Proceedings (PART 2 ed., pp. 73-84). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8404 LNCS, No. PART 2). Springer Verlag. https://doi.org/10.1007/978-3-642-54903-8_7

Yao, Yuanlin ; Xu, Ruifeng ; Lu, Qin et al. / Reader emotion prediction using concept and concept sequence features in news headlines. Computational Linguistics and Intelligent Text Processing - 15th International Conference, CICLing 2014, Proceedings. PART 2. ed. Springer Verlag, 2014. pp. 73-84 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 2).

@inproceedings{08e1a0a6c88d4d0b8363950b26a92a08,

title = "Reader emotion prediction using concept and concept sequence features in news headlines",

abstract = "This paper presents a method to predicate news reader emotions. News headlines supply core information of articles, thus they can serve as key information for reader emotion predication. However, headlines are always short which leads to obvious data sparseness if only lexical forms are used. To address this problem, words in their lexical forms in a headline are transferred to their concepts and concept sequence features of words in headlines based on a semantic knowledge base, namely HowNet for Chinese. These features are expected to represent the major elements which can evoke reader's emotional reactions. These transferred concepts are used with lexical features in headlines for predicating the reader's emotion. Evaluations on dataset of Sina Social News with user emotion votes show that the proposed approach which do not use any news content, achieves a comparable performance to Bag-Of-Word model using both the headlines and the news contents, making our method more efficient in reader emotion prediction.",

keywords = "Concept Feature, Concept Sequence Feature, Emotion Prediction, HowNet",

author = "Yuanlin Yao and Ruifeng Xu and Qin Lu and Bin Liu and Jun Xu and Chengtian Zou and Li Yuan and Shuwei Wang and Lin Yao and Zhenyu He",

year = "2014",

doi = "10.1007/978-3-642-54903-8_7",

language = "English",

isbn = "9783642549021",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

number = "PART 2",

pages = "73--84",

booktitle = "Computational Linguistics and Intelligent Text Processing - 15th International Conference, CICLing 2014, Proceedings",

address = "Germany",

edition = "PART 2",

note = "15th International Conference on Computational Linguistics and Intelligent Text Processing, CICLing 2014 ; Conference date: 06-04-2014 Through 12-04-2014",

}

Yao, Y, Xu, R, Lu, Q, Liu, B, Xu, J, Zou, C, Yuan, L, Wang, S, Yao, L & He, Z 2014, Reader emotion prediction using concept and concept sequence features in news headlines. in Computational Linguistics and Intelligent Text Processing - 15th International Conference, CICLing 2014, Proceedings. PART 2 edn, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 2, vol. 8404 LNCS, Springer Verlag, pp. 73-84, 15th International Conference on Computational Linguistics and Intelligent Text Processing, CICLing 2014, Kathmandu, Nepal, 6/04/14. https://doi.org/10.1007/978-3-642-54903-8_7

Reader emotion prediction using concept and concept sequence features in news headlines. / Yao, Yuanlin; Xu, Ruifeng; Lu, Qin et al.
Computational Linguistics and Intelligent Text Processing - 15th International Conference, CICLing 2014, Proceedings. PART 2. ed. Springer Verlag, 2014. p. 73-84 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8404 LNCS, No. PART 2).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Reader emotion prediction using concept and concept sequence features in news headlines

AU - Yao, Yuanlin

AU - Xu, Ruifeng

AU - Lu, Qin

AU - Liu, Bin

AU - Xu, Jun

AU - Zou, Chengtian

AU - Yuan, Li

AU - Wang, Shuwei

AU - Yao, Lin

AU - He, Zhenyu

PY - 2014

Y1 - 2014

N2 - This paper presents a method to predicate news reader emotions. News headlines supply core information of articles, thus they can serve as key information for reader emotion predication. However, headlines are always short which leads to obvious data sparseness if only lexical forms are used. To address this problem, words in their lexical forms in a headline are transferred to their concepts and concept sequence features of words in headlines based on a semantic knowledge base, namely HowNet for Chinese. These features are expected to represent the major elements which can evoke reader's emotional reactions. These transferred concepts are used with lexical features in headlines for predicating the reader's emotion. Evaluations on dataset of Sina Social News with user emotion votes show that the proposed approach which do not use any news content, achieves a comparable performance to Bag-Of-Word model using both the headlines and the news contents, making our method more efficient in reader emotion prediction.

AB - This paper presents a method to predicate news reader emotions. News headlines supply core information of articles, thus they can serve as key information for reader emotion predication. However, headlines are always short which leads to obvious data sparseness if only lexical forms are used. To address this problem, words in their lexical forms in a headline are transferred to their concepts and concept sequence features of words in headlines based on a semantic knowledge base, namely HowNet for Chinese. These features are expected to represent the major elements which can evoke reader's emotional reactions. These transferred concepts are used with lexical features in headlines for predicating the reader's emotion. Evaluations on dataset of Sina Social News with user emotion votes show that the proposed approach which do not use any news content, achieves a comparable performance to Bag-Of-Word model using both the headlines and the news contents, making our method more efficient in reader emotion prediction.

KW - Concept Feature

KW - Concept Sequence Feature

KW - Emotion Prediction

KW - HowNet

UR - http://www.scopus.com/inward/record.url?scp=84958542334&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-54903-8_7

DO - 10.1007/978-3-642-54903-8_7

M3 - Conference contribution

AN - SCOPUS:84958542334

SN - 9783642549021

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 73

EP - 84

BT - Computational Linguistics and Intelligent Text Processing - 15th International Conference, CICLing 2014, Proceedings

PB - Springer Verlag

T2 - 15th International Conference on Computational Linguistics and Intelligent Text Processing, CICLing 2014

Y2 - 6 April 2014 through 12 April 2014

ER -

Yao Y, Xu R, Lu Q, Liu B, Xu J, Zou C et al. Reader emotion prediction using concept and concept sequence features in news headlines. In Computational Linguistics and Intelligent Text Processing - 15th International Conference, CICLing 2014, Proceedings. PART 2 ed. Springer Verlag. 2014. p. 73-84. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 2). doi: 10.1007/978-3-642-54903-8_7

Reader emotion prediction using concept and concept sequence features in news headlines

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this