A hybrid feature selection method for predicting user influence on Twitter

Yan Mei; Zizhu Zhang; Weiliang Zhao; Jian Yang; Robertus Nugroho

doi:10.1007/978-3-319-26190-4_32

A hybrid feature selection method for predicting user influence on Twitter

Yan Mei^*, Zizhu Zhang, Weiliang Zhao, Jian Yang, Robertus Nugroho

^*Corresponding author for this work

School of Computing

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

Abstract

This paper proposes a hybrid feature selection method for predicting user influence on Twitter. A set of candidate features from Twitter is identified based on the five attributes of influencers defined in sociology. Firstly, less relevant features are filtered out with a featureweighting algorithm. Then the Sequential Backward Floating Selection is utilized as the search strategy. A Back Propagation Neural Network is employed to evaluate the feature subset at each step of searching. Finally, an optimal feature set is obtained for predicting user influence with a high degree of accuracy. Experimental results are provided based on a real world Twitter dataset including seven million tweets associated with 200 popular users. The proposed method can provide a set of features that could be used as a solid foundation for studying complicated user influence evaluation and prediction.

Original language	English
Title of host publication	Web Information Systems Engineering – WISE 2015
Subtitle of host publication	16th International Conference, Miami, FL, USA, November 1-3, 2015, Proceedings, Part 1
Editors	Jianyong Wang, Wojciech Cellary, Dingding Wang, Hua Wang, Shu-Ching Chen, Tao Li, Yanchun Zhang
Place of Publication	Cham
Publisher	Springer, Springer Nature
Pages	478-492
Number of pages	15
Volume	9418
ISBN (Electronic)	9783319261904
ISBN (Print)	9783319261898
DOIs	https://doi.org/10.1007/978-3-319-26190-4_32
Publication status	Published - 2015
Event	16th International Conference on Web Information Systems Engineering, WISE 2015 - Miami, United States Duration: 1 Nov 2015 → 3 Nov 2015

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer International Publishing
Volume	9418
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Other

Other	16th International Conference on Web Information Systems Engineering, WISE 2015
Country/Territory	United States
City	Miami
Period	1/11/15 → 3/11/15

Access to Document

10.1007/978-3-319-26190-4_32

Cite this

Mei, Y., Zhang, Z., Zhao, W., Yang, J., & Nugroho, R. (2015). A hybrid feature selection method for predicting user influence on Twitter. In J. Wang, W. Cellary, D. Wang, H. Wang, S.-C. Chen, T. Li, & Y. Zhang (Eds.), Web Information Systems Engineering – WISE 2015: 16th International Conference, Miami, FL, USA, November 1-3, 2015, Proceedings, Part 1 (Vol. 9418, pp. 478-492). (Lecture Notes in Computer Science; Vol. 9418). Springer, Springer Nature. https://doi.org/10.1007/978-3-319-26190-4_32

Mei, Yan ; Zhang, Zizhu ; Zhao, Weiliang et al. / A hybrid feature selection method for predicting user influence on Twitter. Web Information Systems Engineering – WISE 2015: 16th International Conference, Miami, FL, USA, November 1-3, 2015, Proceedings, Part 1. editor / Jianyong Wang ; Wojciech Cellary ; Dingding Wang ; Hua Wang ; Shu-Ching Chen ; Tao Li ; Yanchun Zhang. Vol. 9418 Cham : Springer, Springer Nature, 2015. pp. 478-492 (Lecture Notes in Computer Science).

@inproceedings{4d300f0394d349d299aaf05aad0266fc,

title = "A hybrid feature selection method for predicting user influence on Twitter",

abstract = "This paper proposes a hybrid feature selection method for predicting user influence on Twitter. A set of candidate features from Twitter is identified based on the five attributes of influencers defined in sociology. Firstly, less relevant features are filtered out with a featureweighting algorithm. Then the Sequential Backward Floating Selection is utilized as the search strategy. A Back Propagation Neural Network is employed to evaluate the feature subset at each step of searching. Finally, an optimal feature set is obtained for predicting user influence with a high degree of accuracy. Experimental results are provided based on a real world Twitter dataset including seven million tweets associated with 200 popular users. The proposed method can provide a set of features that could be used as a solid foundation for studying complicated user influence evaluation and prediction.",

author = "Yan Mei and Zizhu Zhang and Weiliang Zhao and Jian Yang and Robertus Nugroho",

year = "2015",

doi = "10.1007/978-3-319-26190-4_32",

language = "English",

isbn = "9783319261898",

volume = "9418",

series = "Lecture Notes in Computer Science",

publisher = "Springer, Springer Nature",

pages = "478--492",

editor = "Jianyong Wang and Wojciech Cellary and Dingding Wang and Hua Wang and Shu-Ching Chen and Tao Li and Yanchun Zhang",

booktitle = "Web Information Systems Engineering – WISE 2015",

address = "United States",

note = "16th International Conference on Web Information Systems Engineering, WISE 2015 ; Conference date: 01-11-2015 Through 03-11-2015",

}

Mei, Y, Zhang, Z, Zhao, W , Yang, J & Nugroho, R 2015, A hybrid feature selection method for predicting user influence on Twitter. in J Wang, W Cellary, D Wang, H Wang, S-C Chen, T Li & Y Zhang (eds), Web Information Systems Engineering – WISE 2015: 16th International Conference, Miami, FL, USA, November 1-3, 2015, Proceedings, Part 1. vol. 9418, Lecture Notes in Computer Science, vol. 9418, Springer, Springer Nature, Cham, pp. 478-492, 16th International Conference on Web Information Systems Engineering, WISE 2015, Miami, United States, 1/11/15. https://doi.org/10.1007/978-3-319-26190-4_32

A hybrid feature selection method for predicting user influence on Twitter. / Mei, Yan; Zhang, Zizhu; Zhao, Weiliang et al.
Web Information Systems Engineering – WISE 2015: 16th International Conference, Miami, FL, USA, November 1-3, 2015, Proceedings, Part 1. ed. / Jianyong Wang; Wojciech Cellary; Dingding Wang; Hua Wang; Shu-Ching Chen; Tao Li; Yanchun Zhang. Vol. 9418 Cham: Springer, Springer Nature, 2015. p. 478-492 (Lecture Notes in Computer Science; Vol. 9418).

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

TY - GEN

T1 - A hybrid feature selection method for predicting user influence on Twitter

AU - Mei, Yan

AU - Zhang, Zizhu

AU - Zhao, Weiliang

AU - Yang, Jian

AU - Nugroho, Robertus

PY - 2015

Y1 - 2015

N2 - This paper proposes a hybrid feature selection method for predicting user influence on Twitter. A set of candidate features from Twitter is identified based on the five attributes of influencers defined in sociology. Firstly, less relevant features are filtered out with a featureweighting algorithm. Then the Sequential Backward Floating Selection is utilized as the search strategy. A Back Propagation Neural Network is employed to evaluate the feature subset at each step of searching. Finally, an optimal feature set is obtained for predicting user influence with a high degree of accuracy. Experimental results are provided based on a real world Twitter dataset including seven million tweets associated with 200 popular users. The proposed method can provide a set of features that could be used as a solid foundation for studying complicated user influence evaluation and prediction.

AB - This paper proposes a hybrid feature selection method for predicting user influence on Twitter. A set of candidate features from Twitter is identified based on the five attributes of influencers defined in sociology. Firstly, less relevant features are filtered out with a featureweighting algorithm. Then the Sequential Backward Floating Selection is utilized as the search strategy. A Back Propagation Neural Network is employed to evaluate the feature subset at each step of searching. Finally, an optimal feature set is obtained for predicting user influence with a high degree of accuracy. Experimental results are provided based on a real world Twitter dataset including seven million tweets associated with 200 popular users. The proposed method can provide a set of features that could be used as a solid foundation for studying complicated user influence evaluation and prediction.

UR - http://www.scopus.com/inward/record.url?scp=84949949512&partnerID=8YFLogxK

UR - http://purl.org/au-research/grants/arc/LP120200231

U2 - 10.1007/978-3-319-26190-4_32

DO - 10.1007/978-3-319-26190-4_32

M3 - Conference proceeding contribution

AN - SCOPUS:84949949512

SN - 9783319261898

VL - 9418

T3 - Lecture Notes in Computer Science

SP - 478

EP - 492

BT - Web Information Systems Engineering – WISE 2015

A2 - Wang, Jianyong

A2 - Cellary, Wojciech

A2 - Wang, Dingding

A2 - Wang, Hua

A2 - Chen, Shu-Ching

A2 - Li, Tao

A2 - Zhang, Yanchun

PB - Springer, Springer Nature

CY - Cham

T2 - 16th International Conference on Web Information Systems Engineering, WISE 2015

Y2 - 1 November 2015 through 3 November 2015

ER -

Mei Y, Zhang Z, Zhao W , Yang J, Nugroho R. A hybrid feature selection method for predicting user influence on Twitter. In Wang J, Cellary W, Wang D, Wang H, Chen SC, Li T, Zhang Y, editors, Web Information Systems Engineering – WISE 2015: 16th International Conference, Miami, FL, USA, November 1-3, 2015, Proceedings, Part 1. Vol. 9418. Cham: Springer, Springer Nature. 2015. p. 478-492. (Lecture Notes in Computer Science). doi: 10.1007/978-3-319-26190-4_32

A hybrid feature selection method for predicting user influence on Twitter

Abstract

Publication series

Other

Access to Document

Other files and links

Fingerprint

A Personalized Social Network Based Location Search and Recommender System

Cite this

A hybrid feature selection method for predicting user influence on Twitter

Abstract

Publication series

Other

Access to Document

Other files and links

Fingerprint

Projects

A Personalized Social Network Based Location Search and Recommender System

Cite this