This is an open source implementation of the Watset algorithm for fuzzy graph clustering (aka soft clustering). This implementation is multi-threaded, extensible, contains unit tests, and can be easily embedded into any Java-based graph processing pipeline (see examples). Also, it offers a convenient command-line tool for running Watset over the edge lists in text files.

Watset is available on Maven Central and GitHub Packages.

If you publish work that uses Watset, please cite it as follows.

• Ustalov, D., Panchenko, A., Biemann, C., Ponzetto, S.P.: Watset: Local-Global Graph Clustering with Applications in Sense and Frame Induction. Computational Linguistics 45(3), 423–479 (2019)

@article{Ustalov:19:cl,
  author    = {Ustalov, Dmitry and Panchenko, Alexander and Biemann, Chris and Ponzetto, Simone Paolo},
  title     = {{Watset: Local-Global Graph Clustering with Applications in Sense and Frame Induction}},
  journal   = {Computational Linguistics},
  year      = {2019},
  volume    = {45},
  number    = {3},
  pages     = {423--479},
  doi       = {10.1162/COLI_a_00354},
  publisher = {MIT Press},
  issn      = {0891-2017},
  language  = {english},
}

Packages

Package	Description
org.nlpub.watset.cli	Command-line interface.
org.nlpub.watset.eval	Clustering and performance evaluation tools.
org.nlpub.watset.graph	Graph processing and clustering.
org.nlpub.watset.util	Utility classes and helpers.