Influence embedding from incomplete observations in Sina Weibo

Wei Huang; Guohao Sun; Mei Wang; Weiliang Zhao; Jian Yang

doi:10.1007/978-981-99-7254-8_9

Influence embedding from incomplete observations in Sina Weibo

Wei Huang, Guohao Sun^*, Mei Wang, Weiliang Zhao, Jian Yang

^*Corresponding author for this work

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

Abstract

Online Social Networks (OSNs) such as Twitter, Sina Weibo, and Facebook play an important role in our daily life recently. The influence diffusion between users is a common phenomenon on OSNs, which has been applied in numerous applications such as rumor detection and product marketing. Most of the existing influence modeling methods are based on complete data. However, due to certain reasons like privacy protection, it is very hard to obtain complete history data in OSNs. In this paper, we propose a new method to estimate user influence based on incomplete data from user behaviors. Firstly, we apply the maximum likelihood estimator to estimate the user’s missing behaviors. Then, we use direct interaction to get the influence of the sender and receiver. In addition, we apply different actions between users to improve the performance of our method. Empirical experiments on the Weibo dataset show that our method outperforms the existing methods.

Original language	English
Title of host publication	Web Information Systems Engineering – WISE 2023
Subtitle of host publication	24th International Conference, Melbourne, VIC, Australia, October 25–27, 2023, proceedings
Editors	Feng Zhang, Hua Wang, Mahmoud Barhamgi, Lu Chen, Rui Zhou
Place of Publication	Singapore
Publisher	Springer, Springer Nature
Pages	111-121
Number of pages	11
ISBN (Electronic)	9789819972548
ISBN (Print)	9789819972531
DOIs	https://doi.org/10.1007/978-981-99-7254-8_9
Publication status	Published - 2023
Externally published	Yes
Event	24th International Conference on Web Information Systems Engineering, WISE 2023 - Melbourne, Australia Duration: 25 Oct 2023 → 27 Oct 2023

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	14306
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	24th International Conference on Web Information Systems Engineering, WISE 2023
Country/Territory	Australia
City	Melbourne
Period	25/10/23 → 27/10/23

Keywords

Social network
Influence embedding
Sina Weibo

Access to Document

10.1007/978-981-99-7254-8_9

Cite this

Huang, W., Sun, G., Wang, M., Zhao, W., & Yang, J. (2023). Influence embedding from incomplete observations in Sina Weibo. In F. Zhang, H. Wang, M. Barhamgi, L. Chen, & R. Zhou (Eds.), Web Information Systems Engineering – WISE 2023: 24th International Conference, Melbourne, VIC, Australia, October 25–27, 2023, proceedings (pp. 111-121). (Lecture Notes in Computer Science; Vol. 14306). Springer, Springer Nature. https://doi.org/10.1007/978-981-99-7254-8_9

Huang, Wei ; Sun, Guohao ; Wang, Mei et al. / Influence embedding from incomplete observations in Sina Weibo. Web Information Systems Engineering – WISE 2023: 24th International Conference, Melbourne, VIC, Australia, October 25–27, 2023, proceedings. editor / Feng Zhang ; Hua Wang ; Mahmoud Barhamgi ; Lu Chen ; Rui Zhou. Singapore : Springer, Springer Nature, 2023. pp. 111-121 (Lecture Notes in Computer Science).

@inproceedings{4a994057a688430bbba7553a56ed41f0,

title = "Influence embedding from incomplete observations in Sina Weibo",

abstract = "Online Social Networks (OSNs) such as Twitter, Sina Weibo, and Facebook play an important role in our daily life recently. The influence diffusion between users is a common phenomenon on OSNs, which has been applied in numerous applications such as rumor detection and product marketing. Most of the existing influence modeling methods are based on complete data. However, due to certain reasons like privacy protection, it is very hard to obtain complete history data in OSNs. In this paper, we propose a new method to estimate user influence based on incomplete data from user behaviors. Firstly, we apply the maximum likelihood estimator to estimate the user{\textquoteright}s missing behaviors. Then, we use direct interaction to get the influence of the sender and receiver. In addition, we apply different actions between users to improve the performance of our method. Empirical experiments on the Weibo dataset show that our method outperforms the existing methods.",

keywords = "Social network, Influence embedding, Sina Weibo",

author = "Wei Huang and Guohao Sun and Mei Wang and Weiliang Zhao and Jian Yang",

year = "2023",

doi = "10.1007/978-981-99-7254-8_9",

language = "English",

isbn = "9789819972531",

series = "Lecture Notes in Computer Science",

publisher = "Springer, Springer Nature",

pages = "111--121",

editor = "Feng Zhang and Hua Wang and Mahmoud Barhamgi and Lu Chen and Rui Zhou",

booktitle = "Web Information Systems Engineering – WISE 2023",

address = "United States",

note = "24th International Conference on Web Information Systems Engineering, WISE 2023 ; Conference date: 25-10-2023 Through 27-10-2023",

}

Huang, W, Sun, G, Wang, M, Zhao, W & Yang, J 2023, Influence embedding from incomplete observations in Sina Weibo. in F Zhang, H Wang, M Barhamgi, L Chen & R Zhou (eds), Web Information Systems Engineering – WISE 2023: 24th International Conference, Melbourne, VIC, Australia, October 25–27, 2023, proceedings. Lecture Notes in Computer Science, vol. 14306, Springer, Springer Nature, Singapore, pp. 111-121, 24th International Conference on Web Information Systems Engineering, WISE 2023, Melbourne, Australia, 25/10/23. https://doi.org/10.1007/978-981-99-7254-8_9

Influence embedding from incomplete observations in Sina Weibo. / Huang, Wei; Sun, Guohao; Wang, Mei et al.
Web Information Systems Engineering – WISE 2023: 24th International Conference, Melbourne, VIC, Australia, October 25–27, 2023, proceedings. ed. / Feng Zhang; Hua Wang; Mahmoud Barhamgi; Lu Chen; Rui Zhou. Singapore: Springer, Springer Nature, 2023. p. 111-121 (Lecture Notes in Computer Science; Vol. 14306).

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

TY - GEN

T1 - Influence embedding from incomplete observations in Sina Weibo

AU - Huang, Wei

AU - Sun, Guohao

AU - Wang, Mei

AU - Zhao, Weiliang

AU - Yang, Jian

PY - 2023

Y1 - 2023

N2 - Online Social Networks (OSNs) such as Twitter, Sina Weibo, and Facebook play an important role in our daily life recently. The influence diffusion between users is a common phenomenon on OSNs, which has been applied in numerous applications such as rumor detection and product marketing. Most of the existing influence modeling methods are based on complete data. However, due to certain reasons like privacy protection, it is very hard to obtain complete history data in OSNs. In this paper, we propose a new method to estimate user influence based on incomplete data from user behaviors. Firstly, we apply the maximum likelihood estimator to estimate the user’s missing behaviors. Then, we use direct interaction to get the influence of the sender and receiver. In addition, we apply different actions between users to improve the performance of our method. Empirical experiments on the Weibo dataset show that our method outperforms the existing methods.

AB - Online Social Networks (OSNs) such as Twitter, Sina Weibo, and Facebook play an important role in our daily life recently. The influence diffusion between users is a common phenomenon on OSNs, which has been applied in numerous applications such as rumor detection and product marketing. Most of the existing influence modeling methods are based on complete data. However, due to certain reasons like privacy protection, it is very hard to obtain complete history data in OSNs. In this paper, we propose a new method to estimate user influence based on incomplete data from user behaviors. Firstly, we apply the maximum likelihood estimator to estimate the user’s missing behaviors. Then, we use direct interaction to get the influence of the sender and receiver. In addition, we apply different actions between users to improve the performance of our method. Empirical experiments on the Weibo dataset show that our method outperforms the existing methods.

KW - Social network

KW - Influence embedding

KW - Sina Weibo

UR - http://www.scopus.com/inward/record.url?scp=85176011074&partnerID=8YFLogxK

U2 - 10.1007/978-981-99-7254-8_9

DO - 10.1007/978-981-99-7254-8_9

M3 - Conference proceeding contribution

AN - SCOPUS:85176011074

SN - 9789819972531

T3 - Lecture Notes in Computer Science

SP - 111

EP - 121

BT - Web Information Systems Engineering – WISE 2023

A2 - Zhang, Feng

A2 - Wang, Hua

A2 - Barhamgi, Mahmoud

A2 - Chen, Lu

A2 - Zhou, Rui

PB - Springer, Springer Nature

CY - Singapore

T2 - 24th International Conference on Web Information Systems Engineering, WISE 2023

Y2 - 25 October 2023 through 27 October 2023

ER -

Huang W, Sun G, Wang M, Zhao W , Yang J. Influence embedding from incomplete observations in Sina Weibo. In Zhang F, Wang H, Barhamgi M, Chen L, Zhou R, editors, Web Information Systems Engineering – WISE 2023: 24th International Conference, Melbourne, VIC, Australia, October 25–27, 2023, proceedings. Singapore: Springer, Springer Nature. 2023. p. 111-121. (Lecture Notes in Computer Science). doi: 10.1007/978-981-99-7254-8_9