Look before You Leap: Confirming Edge Signs in Random Walk with Restart for Personalized Node Ranking in Signed Networks

Wonchang Lee, Yeon Chang Lee, Dongwon Lee, Sang Wook Kim

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

Abstract

In this paper, we address the personalized node ranking (PNR) problem for signed networks, which aims to rank nodes in an order most relevant to a given seed node in a signed network. The recently-proposed PNR methods introduce the concept of the signed random surfer, denoted as SRSurfer, that performs the score propagation between nodes using the balance theory. However, in real settings of signed networks, edge relationships often do not strictly follow the rules of the balance theory. Therefore, SRSurfer-based PNR methods frequently perform incorrect score propagation to nodes, thereby degrading the accuracy of PNR. To address this limitation, we propose a novel random-walk based PNR approach with sign verification, named as OBOE (lOok Before yOu lEap). Specifically, OBOE carefully verifies the score propagation of SRSurfer by using the topological features of nodes. Then, OBOE corrects all incorrect score propagation cases by exploiting the statistics of a given network. The experiments on 3 real-world signed networks show that OBOE consistently and significantly outperforms 5 competing methods with improvement up to 13%, 95%, and 249% in top-k PNR, bottom-k PNR, and troll identification tasks, respectively. All OBOE codes and datasets are available at: http://github.com/wonchang24/OBOE.

Original languageEnglish (US)
Title of host publicationSIGIR 2021 - Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval
PublisherAssociation for Computing Machinery, Inc
Pages143-152
Number of pages10
ISBN (Electronic)9781450380379
DOIs
StatePublished - Jul 11 2021
Event44th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2021 - Virtual, Online, Canada
Duration: Jul 11 2021Jul 15 2021

Publication series

NameSIGIR 2021 - Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval

Conference

Conference44th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2021
Country/TerritoryCanada
CityVirtual, Online
Period7/11/217/15/21

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Graphics and Computer-Aided Design
  • Information Systems

Fingerprint

Dive into the research topics of 'Look before You Leap: Confirming Edge Signs in Random Walk with Restart for Personalized Node Ranking in Signed Networks'. Together they form a unique fingerprint.

Cite this