Multi-label Learning with a Cone-Based Geometric Model

Mena Leemhuis*, Özgür L. Özçep, Diedrich Wolter

*Corresponding author for this work


Recent approaches for knowledge-graph embeddings aim at connecting quantitative data structures used in machine learning to the qualitative structures of logics. Such embeddings are of a hybrid nature, they are data models that also exhibit conceptual structures inherent to logics. One motivation to investigate embeddings is to design conceptually adequate machine learning (ML) algorithms. This paper investigates a new approach to embedding ontologies into geometric models that interpret concepts by closed convex cones. As a proof of concept this cone-based embedding was implemented in a ML algorithm for weak supervised multi-label learning. The system was tested with the gene ontology and showed a performance similar to comparable approaches, but with the advantage of exhibiting the conceptual structure underlying the data.

Original languageEnglish
Title of host publicationICCS 2020: Ontologies and Concepts in Mind and Machine
EditorsMehwish Alam, Tanya Braun, Bruno Yun
Number of pages9
Volume12277 LNAI
PublisherSpringer, Cham
Publication date10.09.2020
ISBN (Print)978-3-030-57854-1
ISBN (Electronic)978-3-030-57855-8
Publication statusPublished - 10.09.2020
Event25th International Conference on Conceptual Structures - Bolzano, Italy
Duration: 18.09.202020.09.2020
Conference number: 245219

Research Areas and Centers

  • Centers: Center for Artificial Intelligence Luebeck (ZKIL)
  • Research Area: Intelligent Systems


Dive into the research topics of 'Multi-label Learning with a Cone-Based Geometric Model'. Together they form a unique fingerprint.

Cite this