Which Group Do You Belong To? Sentiment-Based PageRank to Measure Formal and Informal Influence of Nodes in Networks

Lan Jiang, Ly Dinh, Rezvaneh Rezapour, Jana Diesner

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


Organizational networks are often hierarchical by nature as individuals take on roles or functions at various job levels. Prior studies have used either text-level (e.g., sentiment, affect) or structural-level features (e.g., PageRank, various centrality metrics) to identify influential nodes in networks. In this study, we use a combination of these two levels of information to develop a novel ranking method that combines sentiment analysis and PageRank to infer node-level influence in a real-world organizational network. We detect sentiment scores for all actor pairs based on the content of their email-based communication, and calculate their influence index using an enhanced PageRank method. Finally, we group individual nodes into distinct clusters according to their influence index. Compared to established network metrics designed or used to infer formal and informal influence and ground truth data on job levels, our metric achieves the highest accuracy for inferring formal influence (60.7%) and second highest for inferring informal influence (69.0%). Our approach shows that combining text-level and structural-level information is effective for identifying the job level of nodes in an organizational network.

Original languageEnglish (US)
Title of host publicationComplex Networks and Their Applications IX - Volume 2, Proceedings of the Ninth International Conference on Complex Networks and Their Applications COMPLEX NETWORKS 2020
EditorsRosa M. Benito, Chantal Cherifi, Hocine Cherifi, Esteban Moro, Luis Mateus Rocha, Marta Sales-Pardo
Number of pages14
ISBN (Print)9783030653507
StatePublished - 2021
Event9th International Conference on Complex Networks and their Applications, COMPLEX NETWORKS 2020 - Madrid, Spain
Duration: Dec 1 2020Dec 3 2020

Publication series

NameStudies in Computational Intelligence
ISSN (Print)1860-949X
ISSN (Electronic)1860-9503


Conference9th International Conference on Complex Networks and their Applications, COMPLEX NETWORKS 2020


  • Formal influence
  • Informal influence
  • Organizational networks
  • PageRank
  • Sentiment analysis

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this