Abstract

Users' personal information such as their political views is important for many applications such as targeted advertisements or real-time monitoring of political opinions. Huge amounts of data generated by social media users present opportunities and challenges to study these preferences in a large scale. In this paper, we aim to infer social media users' political views when only network information is available. In particular, given personal preferences about some of the social media users, how can we infer the preferences of unobserved individuals in the same network? There are many existing solutions that address the problem of classification with networked data problem. However, networks in social media normally involve millions and even hundreds of millions of nodes, which make the scalability an important problem in inferring personal preferences in social media. To address the scalability issue, we use social influence theory to construct new features based on a combination of local and global structures of the network. Then we use these features to train classifiers and predict users' preferences. Due to the size of real-world social networks, using the entire network information is inefficient and not practical in many cases. By extracting local social dimensions, we present an efficient and scalable solution. Further, by capturing the network's global pattern, the proposed solution, balances the performance requirement between accuracy and efficiency.

Original languageEnglish (US)
Title of host publicationHT 2014 - Proceedings of the 25th ACM Conference on Hypertext and Social Media
PublisherAssociation for Computing Machinery
Pages4-12
Number of pages9
ISBN (Print)9781450329545
DOIs
StatePublished - 2014
Event25th ACM Conference on Hypertext and Social Media, HT 2014 - Santiago, Chile
Duration: Sep 1 2014Sep 4 2014

Other

Other25th ACM Conference on Hypertext and Social Media, HT 2014
CountryChile
CitySantiago
Period9/1/149/4/14

Fingerprint

Scalability
Classifiers
Monitoring

Keywords

  • homophily
  • preference prediction
  • relational learning
  • social media mining

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Human-Computer Interaction
  • Software

Cite this

Abbasi, M. A., Tang, J., & Liu, H. (2014). Scalable learning of users' preferences using networked data. In HT 2014 - Proceedings of the 25th ACM Conference on Hypertext and Social Media (pp. 4-12). Association for Computing Machinery. https://doi.org/10.1145/2631775.2631796

Scalable learning of users' preferences using networked data. / Abbasi, Mohammad Ali; Tang, Jiliang; Liu, Huan.

HT 2014 - Proceedings of the 25th ACM Conference on Hypertext and Social Media. Association for Computing Machinery, 2014. p. 4-12.

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

Abbasi, MA, Tang, J & Liu, H 2014, Scalable learning of users' preferences using networked data. in HT 2014 - Proceedings of the 25th ACM Conference on Hypertext and Social Media. Association for Computing Machinery, pp. 4-12, 25th ACM Conference on Hypertext and Social Media, HT 2014, Santiago, Chile, 9/1/14. https://doi.org/10.1145/2631775.2631796
Abbasi MA, Tang J, Liu H. Scalable learning of users' preferences using networked data. In HT 2014 - Proceedings of the 25th ACM Conference on Hypertext and Social Media. Association for Computing Machinery. 2014. p. 4-12 https://doi.org/10.1145/2631775.2631796
Abbasi, Mohammad Ali ; Tang, Jiliang ; Liu, Huan. / Scalable learning of users' preferences using networked data. HT 2014 - Proceedings of the 25th ACM Conference on Hypertext and Social Media. Association for Computing Machinery, 2014. pp. 4-12
@inproceedings{8877f2aa653e4ce39ec7343550cc2143,
title = "Scalable learning of users' preferences using networked data",
abstract = "Users' personal information such as their political views is important for many applications such as targeted advertisements or real-time monitoring of political opinions. Huge amounts of data generated by social media users present opportunities and challenges to study these preferences in a large scale. In this paper, we aim to infer social media users' political views when only network information is available. In particular, given personal preferences about some of the social media users, how can we infer the preferences of unobserved individuals in the same network? There are many existing solutions that address the problem of classification with networked data problem. However, networks in social media normally involve millions and even hundreds of millions of nodes, which make the scalability an important problem in inferring personal preferences in social media. To address the scalability issue, we use social influence theory to construct new features based on a combination of local and global structures of the network. Then we use these features to train classifiers and predict users' preferences. Due to the size of real-world social networks, using the entire network information is inefficient and not practical in many cases. By extracting local social dimensions, we present an efficient and scalable solution. Further, by capturing the network's global pattern, the proposed solution, balances the performance requirement between accuracy and efficiency.",
keywords = "homophily, preference prediction, relational learning, social media mining",
author = "Abbasi, {Mohammad Ali} and Jiliang Tang and Huan Liu",
year = "2014",
doi = "10.1145/2631775.2631796",
language = "English (US)",
isbn = "9781450329545",
pages = "4--12",
booktitle = "HT 2014 - Proceedings of the 25th ACM Conference on Hypertext and Social Media",
publisher = "Association for Computing Machinery",

}

TY - GEN

T1 - Scalable learning of users' preferences using networked data

AU - Abbasi, Mohammad Ali

AU - Tang, Jiliang

AU - Liu, Huan

PY - 2014

Y1 - 2014

N2 - Users' personal information such as their political views is important for many applications such as targeted advertisements or real-time monitoring of political opinions. Huge amounts of data generated by social media users present opportunities and challenges to study these preferences in a large scale. In this paper, we aim to infer social media users' political views when only network information is available. In particular, given personal preferences about some of the social media users, how can we infer the preferences of unobserved individuals in the same network? There are many existing solutions that address the problem of classification with networked data problem. However, networks in social media normally involve millions and even hundreds of millions of nodes, which make the scalability an important problem in inferring personal preferences in social media. To address the scalability issue, we use social influence theory to construct new features based on a combination of local and global structures of the network. Then we use these features to train classifiers and predict users' preferences. Due to the size of real-world social networks, using the entire network information is inefficient and not practical in many cases. By extracting local social dimensions, we present an efficient and scalable solution. Further, by capturing the network's global pattern, the proposed solution, balances the performance requirement between accuracy and efficiency.

AB - Users' personal information such as their political views is important for many applications such as targeted advertisements or real-time monitoring of political opinions. Huge amounts of data generated by social media users present opportunities and challenges to study these preferences in a large scale. In this paper, we aim to infer social media users' political views when only network information is available. In particular, given personal preferences about some of the social media users, how can we infer the preferences of unobserved individuals in the same network? There are many existing solutions that address the problem of classification with networked data problem. However, networks in social media normally involve millions and even hundreds of millions of nodes, which make the scalability an important problem in inferring personal preferences in social media. To address the scalability issue, we use social influence theory to construct new features based on a combination of local and global structures of the network. Then we use these features to train classifiers and predict users' preferences. Due to the size of real-world social networks, using the entire network information is inefficient and not practical in many cases. By extracting local social dimensions, we present an efficient and scalable solution. Further, by capturing the network's global pattern, the proposed solution, balances the performance requirement between accuracy and efficiency.

KW - homophily

KW - preference prediction

KW - relational learning

KW - social media mining

UR - http://www.scopus.com/inward/record.url?scp=84907420692&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84907420692&partnerID=8YFLogxK

U2 - 10.1145/2631775.2631796

DO - 10.1145/2631775.2631796

M3 - Conference contribution

AN - SCOPUS:84907420692

SN - 9781450329545

SP - 4

EP - 12

BT - HT 2014 - Proceedings of the 25th ACM Conference on Hypertext and Social Media

PB - Association for Computing Machinery

ER -