ContrastMedium algorithm
Stefano Faralli1 Alexander Panchenko2 Chris Biemann2 Simone Paolo Ponzetto1
1 Data and Web Science Group, University of Mannheim, Germany
{stefano,simone}@informatik.uni-mannheim.de
2 Language Technology Group, University of Hamburg, Germany
{panchenko,biemann}@informatik.uni-hamburg.de

 

We present ContrastMedium (CM), an algorithm that transforms noisy semantic networks into full-fledged, clean taxonomies. CM is able to identify the embedded taxonomy structure from a noisy knowledge graph without explicit human supervision, e.g., a set of manually selected input root and leaf concepts. This is achieved by leveraging structural information from a companion reference taxonomy, to which the input knowledge graph is linked (either automatically or manually). When used in conjunction with un-supervised methods for hypernym acquisition and knowledge base linking, our methodology provides a complete solution for end-to-end taxonomy induction. Our experiments using CM on automatically acquired knowledge graphs, as well as a SemEval benchmark, achieved good re sults.

 

- Interactive examples:
We provide some interactive examples: example 1, example 2, example 3, example 4, example 5, example 6,
example 7, example 8, example 9, example 10

 

- Source code:

GitHub repository of ContrastMedium as a NetBeans java application project.
 

- Publications:

- Stefano Faralli, Alexander Panchenko, Chris Biemann and Simone Paolo Ponzetto (2017) The ContrastMedium algorithm : taxonomy induction from noisy knowledge graphs with just a few links. 15th Conference of the European Chapter of the Association for Computational Linguistics : proceedings of conference, volume 1: Long Papers; 590-600. Association for Computational Linguistics. bib

- License

All the datasets and software are licensed under a Creative Commons Attribution 4.0 International (CC BY 4.0) License: https://creativecommons.org/licenses/by/4.0/.


DFG project JOIN-T