Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
September 29, 2025Open Access

Clustering scientific publications: lessons learned through experiments with a real citation network

View Full Paper
Ask AI
Bookmark
Share

Authors

VHVũ Thị Mai HườngTKThorsten Koch

Discussion

Loading...

Member takes

Overview

This analysis tests Louvain and Leiden clustering methods on a citation network, highlighting parameter tuning needs for accurate outcomes.

Key Points

  • Results demonstrate that scalable methods like Louvain and Leiden can fail without proper settings.
  • Analysis was conducted on a citation graph of approximately 700,000 papers and 4.6 million citations.
  • Graph-based clustering methods are dependent on careful parameter tuning for effective partitioning.
  • Findings emphasize practical challenges in clustering large-scale bibliometric data and structure-specific method selection.

Cite This Study

Hường et al. (2025) studied this question.

synapsesocial.com/papers/68da58d8c1728099cfd10fe7https://doi.org/10.48550/arxiv.2505.18180
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1ClusterChoice: a web-based tool for interpretable clustering and algorithm comparison in bibliometric networks2026 · 1 citations
  2. 2Developing a novel algorithm for comparing cluster patterns in networks on journal articles during and after COVID-19: Bibliometric analysis2024 · 1 citations
  3. 3Clustering Scientific Document Based on an Extended Citation Model2019 · 10 citations
  4. 4Enhancing Bibliometric Insights: A Novel Overlapping Clustering Framework for Citation Networks2025
  5. 5Exploring Citation and Author Collaboration Network with Graph Database2024