<?xml version="1.0" encoding="utf-8"?>
<export-example>
  <doc>
    <id>10041</id>
    <completedYear/>
    <publishedYear/>
    <thesisYearAccepted/>
    <language>eng</language>
    <pageFirst/>
    <pageLast/>
    <pageNumber/>
    <edition/>
    <issue/>
    <volume/>
    <type>reportzib</type>
    <publisherName/>
    <publisherPlace/>
    <creatingCorporation/>
    <contributingCorporation/>
    <belongsToBibliography>0</belongsToBibliography>
    <completedDate>--</completedDate>
    <publishedDate>2025-06-05</publishedDate>
    <thesisDateAccepted>--</thesisDateAccepted>
    <title language="eng">Clustering scientific publications: lessons learned through experiments with a real citation network</title>
    <abstract language="eng">Clustering scientific publications helps uncover research structures within bibliographic databases. Graph-based methods such as spectral, Louvain, and Leiden clustering are commonly used due to their ability to model citation networks. However, their effectiveness can diminish when applied to real-world data. This study evaluates these clustering algorithms on a citation graph of about 700,000 articles and 4.6 million citations from the Web of Science. The results show that while scalable methods like Louvain and Leiden perform efficiently, their default settings often yield poor partitioning. Meaningful outcomes require careful parameter tuning, especially for large networks with uneven structures, including a dense core and loosely connected papers. These findings highlight practical lessons about the challenges of large-scale data, method selection and tuning based on specific structures of bibliometric clustering tasks.</abstract>
    <identifier type="arxiv">2505.18180</identifier>
    <identifier type="urn">urn:nbn:de:0297-zib-100418</identifier>
    <enrichment key="opus.source">publish</enrichment>
    <author>Thi Huong Vu</author>
    <submitter>Thi Huong Vu</submitter>
    <author>Thorsten Koch</author>
    <series>
      <title>ZIB-Report</title>
      <number>25-05</number>
    </series>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>graph clustering</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>citation networks</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>Web of Science</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>bibliometric analysis</value>
    </subject>
    <subject>
      <language>eng</language>
      <type>uncontrolled</type>
      <value>unsupervised learning</value>
    </subject>
    <collection role="institutes" number="sis">Digital Data and Information for Society, Science, and Culture</collection>
    <collection role="persons" number="koch">Koch, Thorsten</collection>
    <collection role="institutes" number="aim">Applied Algorithmic Intelligence Methods</collection>
    <collection role="persons" number="huong.vu">Vu, Thi Huong</collection>
    <collection role="projects" number="FAN">FAN</collection>
    <collection role="institutes" number="aopt">Applied Optimization</collection>
    <file>https://opus4.kobv.de/opus4-zib/files/10041/main.pdf</file>
  </doc>
</export-example>
