Towards large-scale social networks with online diffusion provenance detection

Haishuai Wang, Jia Wu, Shirui Pan, Peng Zhang, Ling Chen

Research output: Contribution to journalArticleResearchpeer-review

Abstract

In this paper we study a new problem of online discovering diffusion provenances in large networks. Existing work on network diffusion provenance identification focuses on offline learning where data collected from network detectors are static and a snapshot of the network is available before learning. However, an offline learning model does not meet the need for early warning, real-time awareness, or a real-time response to malicious information spreading in networks. To this end, we propose an online regression model for real-time diffusion provenance identification. Specifically, we first use offline collected network cascades to infer the edge transmission weights, and then use an online l1 non-convex regression model as the identification model. The proposed methods are empirically evaluated on both synthetic and real-world networks. Experimental results demonstrate the effectiveness of the proposed model.

Original languageEnglish
Pages (from-to)154-166
Number of pages13
JournalComputer Networks
Volume114
DOIs
Publication statusPublished - 26 Feb 2017
Externally publishedYes

Keywords

  • L Regression
  • Online identification
  • Social network
  • Source locating

Cite this

Wang, Haishuai ; Wu, Jia ; Pan, Shirui ; Zhang, Peng ; Chen, Ling. / Towards large-scale social networks with online diffusion provenance detection. In: Computer Networks. 2017 ; Vol. 114. pp. 154-166.
@article{3009ff3309f84c7fb2bcd084b807f94c,
title = "Towards large-scale social networks with online diffusion provenance detection",
abstract = "In this paper we study a new problem of online discovering diffusion provenances in large networks. Existing work on network diffusion provenance identification focuses on offline learning where data collected from network detectors are static and a snapshot of the network is available before learning. However, an offline learning model does not meet the need for early warning, real-time awareness, or a real-time response to malicious information spreading in networks. To this end, we propose an online regression model for real-time diffusion provenance identification. Specifically, we first use offline collected network cascades to infer the edge transmission weights, and then use an online l1 non-convex regression model as the identification model. The proposed methods are empirically evaluated on both synthetic and real-world networks. Experimental results demonstrate the effectiveness of the proposed model.",
keywords = "L Regression, Online identification, Social network, Source locating",
author = "Haishuai Wang and Jia Wu and Shirui Pan and Peng Zhang and Ling Chen",
year = "2017",
month = "2",
day = "26",
doi = "10.1016/j.comnet.2016.08.025",
language = "English",
volume = "114",
pages = "154--166",
journal = "Computer Networks",
issn = "1389-1286",
publisher = "Elsevier",

}

Towards large-scale social networks with online diffusion provenance detection. / Wang, Haishuai; Wu, Jia; Pan, Shirui; Zhang, Peng; Chen, Ling.

In: Computer Networks, Vol. 114, 26.02.2017, p. 154-166.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Towards large-scale social networks with online diffusion provenance detection

AU - Wang, Haishuai

AU - Wu, Jia

AU - Pan, Shirui

AU - Zhang, Peng

AU - Chen, Ling

PY - 2017/2/26

Y1 - 2017/2/26

N2 - In this paper we study a new problem of online discovering diffusion provenances in large networks. Existing work on network diffusion provenance identification focuses on offline learning where data collected from network detectors are static and a snapshot of the network is available before learning. However, an offline learning model does not meet the need for early warning, real-time awareness, or a real-time response to malicious information spreading in networks. To this end, we propose an online regression model for real-time diffusion provenance identification. Specifically, we first use offline collected network cascades to infer the edge transmission weights, and then use an online l1 non-convex regression model as the identification model. The proposed methods are empirically evaluated on both synthetic and real-world networks. Experimental results demonstrate the effectiveness of the proposed model.

AB - In this paper we study a new problem of online discovering diffusion provenances in large networks. Existing work on network diffusion provenance identification focuses on offline learning where data collected from network detectors are static and a snapshot of the network is available before learning. However, an offline learning model does not meet the need for early warning, real-time awareness, or a real-time response to malicious information spreading in networks. To this end, we propose an online regression model for real-time diffusion provenance identification. Specifically, we first use offline collected network cascades to infer the edge transmission weights, and then use an online l1 non-convex regression model as the identification model. The proposed methods are empirically evaluated on both synthetic and real-world networks. Experimental results demonstrate the effectiveness of the proposed model.

KW - L Regression

KW - Online identification

KW - Social network

KW - Source locating

UR - http://www.scopus.com/inward/record.url?scp=84994750509&partnerID=8YFLogxK

U2 - 10.1016/j.comnet.2016.08.025

DO - 10.1016/j.comnet.2016.08.025

M3 - Article

VL - 114

SP - 154

EP - 166

JO - Computer Networks

JF - Computer Networks

SN - 1389-1286

ER -