A network-based model for predicting hashtag breakouts in twitter

Sultan Alzahrani; Saud Alashri; Anvesh Reddy Koppela; Hasan Davulcu; Ismail Toroslu

doi:10.1007/978-3-319-16268-3_1

A network-based model for predicting hashtag breakouts in twitter

Sultan Alzahrani, Saud Alashri, Anvesh Reddy Koppela, Hasan Davulcu, Ismail Toroslu

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

3 Scopus citations

Abstract

Online information propagates differently on the web, some of which can be viral. In this paper, first we introduce a simple standard deviation sigma levels based Tweet volume breakout definition, then we proceed to determine patterns of re-tweet network measures to predict whether a hashtag volume will breakout or not. We also developed a visualization tool to help trace the evolution of hashtag volumes, their underlying networks and both local and global network measures. We trained a random forest tree classifier to identify effective network measures for predicting hashtag volume breakouts. Our experiments showed that “local” network features, based on a fixed-sized sliding window, have an overall predictive accuracy of 76%, where as, when we incorporate “global” features that utilize all interactions up to the current period, then the overall predictive accuracy of a sliding window based breakout predictor jumps to 83%.

Original language	English (US)
Title of host publication	Social Computing, Behavioral-Cultural Modeling, and Prediction - 8th International Conference, SBP 2015, Proceedings
Editors	Kevin Xu, Nitin Agarwal, Nathaniel Osgood
Publisher	Springer Verlag
Pages	3-12
Number of pages	10
ISBN (Electronic)	9783319162676
DOIs	https://doi.org/10.1007/978-3-319-16268-3_1
State	Published - 2015
Event	8th International Conference on Social Computing, Behavioral-Cultural Modeling, and Prediction, SBP 2015 - Washington, United States Duration: Mar 31 2015 → Apr 3 2015

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	9021
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Other

Other	8th International Conference on Social Computing, Behavioral-Cultural Modeling, and Prediction, SBP 2015
Country/Territory	United States
City	Washington
Period	3/31/15 → 4/3/15

Keywords

Diffusion networks
Hashtag volumes
Information diffusion
Prediction
Social networks

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-319-16268-3_1

Cite this

Alzahrani, S., Alashri, S., Koppela, A. R., Davulcu, H., & Toroslu, I. (2015). A network-based model for predicting hashtag breakouts in twitter. In K. Xu, N. Agarwal, & N. Osgood (Eds.), Social Computing, Behavioral-Cultural Modeling, and Prediction - 8th International Conference, SBP 2015, Proceedings (pp. 3-12). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9021). Springer Verlag. https://doi.org/10.1007/978-3-319-16268-3_1

A network-based model for predicting hashtag breakouts in twitter. / Alzahrani, Sultan; Alashri, Saud; Koppela, Anvesh Reddy et al.
Social Computing, Behavioral-Cultural Modeling, and Prediction - 8th International Conference, SBP 2015, Proceedings. ed. / Kevin Xu; Nitin Agarwal; Nathaniel Osgood. Springer Verlag, 2015. p. 3-12 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9021).

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

Alzahrani, S, Alashri, S, Koppela, AR, Davulcu, H & Toroslu, I 2015, A network-based model for predicting hashtag breakouts in twitter. in K Xu, N Agarwal & N Osgood (eds), Social Computing, Behavioral-Cultural Modeling, and Prediction - 8th International Conference, SBP 2015, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9021, Springer Verlag, pp. 3-12, 8th International Conference on Social Computing, Behavioral-Cultural Modeling, and Prediction, SBP 2015, Washington, United States, 3/31/15. https://doi.org/10.1007/978-3-319-16268-3_1

Alzahrani S, Alashri S, Koppela AR, Davulcu H, Toroslu I. A network-based model for predicting hashtag breakouts in twitter. In Xu K, Agarwal N, Osgood N, editors, Social Computing, Behavioral-Cultural Modeling, and Prediction - 8th International Conference, SBP 2015, Proceedings. Springer Verlag. 2015. p. 3-12. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-16268-3_1

Alzahrani, Sultan ; Alashri, Saud ; Koppela, Anvesh Reddy et al. / A network-based model for predicting hashtag breakouts in twitter. Social Computing, Behavioral-Cultural Modeling, and Prediction - 8th International Conference, SBP 2015, Proceedings. editor / Kevin Xu ; Nitin Agarwal ; Nathaniel Osgood. Springer Verlag, 2015. pp. 3-12 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{f19d907ce59f4db5883a33172ef7ea98,

title = "A network-based model for predicting hashtag breakouts in twitter",

abstract = "Online information propagates differently on the web, some of which can be viral. In this paper, first we introduce a simple standard deviation sigma levels based Tweet volume breakout definition, then we proceed to determine patterns of re-tweet network measures to predict whether a hashtag volume will breakout or not. We also developed a visualization tool to help trace the evolution of hashtag volumes, their underlying networks and both local and global network measures. We trained a random forest tree classifier to identify effective network measures for predicting hashtag volume breakouts. Our experiments showed that “local” network features, based on a fixed-sized sliding window, have an overall predictive accuracy of 76%, where as, when we incorporate “global” features that utilize all interactions up to the current period, then the overall predictive accuracy of a sliding window based breakout predictor jumps to 83%.",

keywords = "Diffusion networks, Hashtag volumes, Information diffusion, Prediction, Social networks",

author = "Sultan Alzahrani and Saud Alashri and Koppela, {Anvesh Reddy} and Hasan Davulcu and Ismail Toroslu",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing Switzerland 2015.; 8th International Conference on Social Computing, Behavioral-Cultural Modeling, and Prediction, SBP 2015 ; Conference date: 31-03-2015 Through 03-04-2015",

year = "2015",

doi = "10.1007/978-3-319-16268-3_1",

language = "English (US)",

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

publisher = "Springer Verlag",

pages = "3--12",

editor = "Kevin Xu and Nitin Agarwal and Nathaniel Osgood",

booktitle = "Social Computing, Behavioral-Cultural Modeling, and Prediction - 8th International Conference, SBP 2015, Proceedings",

}

TY - GEN

T1 - A network-based model for predicting hashtag breakouts in twitter

AU - Alzahrani, Sultan

AU - Alashri, Saud

AU - Koppela, Anvesh Reddy

AU - Davulcu, Hasan

AU - Toroslu, Ismail

N1 - Publisher Copyright: © Springer International Publishing Switzerland 2015.

PY - 2015

Y1 - 2015

N2 - Online information propagates differently on the web, some of which can be viral. In this paper, first we introduce a simple standard deviation sigma levels based Tweet volume breakout definition, then we proceed to determine patterns of re-tweet network measures to predict whether a hashtag volume will breakout or not. We also developed a visualization tool to help trace the evolution of hashtag volumes, their underlying networks and both local and global network measures. We trained a random forest tree classifier to identify effective network measures for predicting hashtag volume breakouts. Our experiments showed that “local” network features, based on a fixed-sized sliding window, have an overall predictive accuracy of 76%, where as, when we incorporate “global” features that utilize all interactions up to the current period, then the overall predictive accuracy of a sliding window based breakout predictor jumps to 83%.

AB - Online information propagates differently on the web, some of which can be viral. In this paper, first we introduce a simple standard deviation sigma levels based Tweet volume breakout definition, then we proceed to determine patterns of re-tweet network measures to predict whether a hashtag volume will breakout or not. We also developed a visualization tool to help trace the evolution of hashtag volumes, their underlying networks and both local and global network measures. We trained a random forest tree classifier to identify effective network measures for predicting hashtag volume breakouts. Our experiments showed that “local” network features, based on a fixed-sized sliding window, have an overall predictive accuracy of 76%, where as, when we incorporate “global” features that utilize all interactions up to the current period, then the overall predictive accuracy of a sliding window based breakout predictor jumps to 83%.

KW - Diffusion networks

KW - Hashtag volumes

KW - Information diffusion

KW - Prediction

KW - Social networks

UR - http://www.scopus.com/inward/record.url?scp=84925303347&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84925303347&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-16268-3_1

DO - 10.1007/978-3-319-16268-3_1

M3 - Conference contribution

AN - SCOPUS:84925303347

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

SP - 3

EP - 12

BT - Social Computing, Behavioral-Cultural Modeling, and Prediction - 8th International Conference, SBP 2015, Proceedings

A2 - Xu, Kevin

A2 - Agarwal, Nitin

A2 - Osgood, Nathaniel

PB - Springer Verlag

T2 - 8th International Conference on Social Computing, Behavioral-Cultural Modeling, and Prediction, SBP 2015

Y2 - 31 March 2015 through 3 April 2015

ER -

A network-based model for predicting hashtag breakouts in twitter

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this