Online community conflict decomposition with pseudo spatial permutation

Yunmo Chen, Xinyue Ye

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

1 Scopus citations


Online communities are composed of individuals sharing similar opinions or behavior in the virtual world. Facilitated by the fast development of social media platforms, the expansion of online communities have raised many attentions among the researchers, business analysts, and decision makers, leading to a growing list of literature studying the interactions especially conflicts in the online communities. A conflict is often initiated by one community which then attacks the other, leading to an adversarial relationship and worse social impacts. Many studies have examined the origins and process of online community conflict while failing to address the possible spatial effects in their models. In this paper, we explore the prediction of online community conflict by decomposing and analyzing its prediction error taking geography into accounts. Grounding on the previous natural language processing based model, we introduce pseudo spatial permutation to test the model expressiveness with geographical factors. Pseudo spatial permutation employs different geographical distributions to sample from and perturbs the model using the pseudo geographical information to examine the relationship between online community conflict and spatial distribution. Our analysis shows that the pseudo spatial permutation is an efficient approach to robustly test the conflict relation learned by the prediction model, and also reveals the necessity to incorporate geographical information into the prediction. In conclusion, this work provides a different aspect of analyzing the community conflict that does not solely rely on the textual communication.

Original languageEnglish (US)
Title of host publicationComputational Data and Social Networks - 8th International Conference, CSoNet 2019, Proceedings
EditorsAndrea Tagarelli, Hanghang Tong
Number of pages10
ISBN (Print)9783030349790
StatePublished - 2019
Event8th International Conference on Computational Data and Social Networks, CSoNet 2019 - Ho Chi Minh City, Viet Nam
Duration: Nov 18 2019Nov 20 2019

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11917 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference8th International Conference on Computational Data and Social Networks, CSoNet 2019
Country/TerritoryViet Nam
CityHo Chi Minh City

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • General Computer Science


  • Neural network
  • Online community
  • Spatial permutation
  • Spatial social network
  • Text mining


Dive into the research topics of 'Online community conflict decomposition with pseudo spatial permutation'. Together they form a unique fingerprint.

Cite this