Jens Lehmann

Concept Learning in Description Logics

The problem of learning logic programs has been researched extensively, but other knowledge representation formalisms like Description Logics are also an interesting target language. The importance of inductive reasoning in Description Logics has increased with the rise of the Semantic Web, because the learning algorithms can be used as a means for the computer aided building of ontologies. Ontology construction is a burdensome task and powerful tools are needed to support knowledge engineers.

One of the keys for designing induction algorithms in Description Logics are refinement operators. They allow for an efficient traversal of the subsumption hierarchy of concepts. One way to assess the suitability of a refinement operator for learning algorithms is to look at its properties. We analysed the properties completeness, weak completeness, properness, redundancy, finiteness, minimality, and their combinations, in particular we show theoretical limitations of refinement operators in Descriptions Logics.

Learning algorithms can be designed by combining a refinement operator with a a search heuristic. We propose an operator and show that it is close to the best we can hope for. We then create a sound learning algorithm by adding an intelligent search heuristic.

As a second approach we investigate the use of Genetic Programming to solve the learning problem in Description Logics. We discuss the characteristics of Genetic Programming in this context and show a way to incorporate refinement operators in the Genetic Programming framework. Again, we define a suitable operator and analyse it. Some further extensions are also proposed.

Journal Publications

• "DL-Learner: Learning Concepts in Description Logics"
  Jens Lehmann
  Journal of Machine Learning Research (JMLR), 2009
  
  • download: PDF (113.4 KB)
  • BibTeX entry: BibTeX entry
  • related info pages: Concept Learning in Description Logics
  • related weblinks: DL-Learner homepage
  
  
• "Concept Learning in Description Logics Using Refinement Operators"
  Jens Lehmann, Pascal Hitzler
  Machine Learning, , volume 78, pages 203-250, Springer, 2010
  
  • download: PDF (403.9 KB)
  • BibTeX entry: BibTeX entry
  • related info pages: Concept Learning in Description Logics
  • related weblinks: DL-Learner homepage
  
  
• "Learning of OWL Class Descriptions on Very Large Knowledge Bases"
  Sebastian Hellmann, Jens Lehmann, Sören Auer
  International Journal On Semantic Web and Information Systems, 2009
  
  • download: PDF (447.6 KB)
  • BibTeX entry: BibTeX entry
  • related info pages: Concept Learning in Description Logics
  • related weblinks: DBpedia Homepage, DL-Learner homepage
  
  

Conference Publications

• "A Refinement Operator Based Learning Algorithm for the ALC Description Logic"
  Jens Lehmann, Pascal Hitzler
  In: Hendrick Blockeel, Jude W. Shavlik, Prasad Tadepalli (editors), Proceedings of the 17th International Conference on Inductive Logic Programming (ILP), Lecture Notes in Computer Science, volume 4894, pages 147--160, Springer, 2008
  
  • download: PDF (240.1 KB)
  • BibTeX entry: BibTeX entry
  • related info pages: Concept Learning in Description Logics
  • Best Student Paper Award ($1500 sponsored by the Machine Learning Journal, shared with the paper below)
  
  
• "Foundations of Refinement Operators for Description Logics"
  Jens Lehmann, Pascal Hitzler
  In: Hendrick Blockeel, Jude W. Shavlik, Prasad Tadepalli (editors), Proceedings of the 17th International Conference on Inductive Logic Programming (ILP), Lecture Notes in Computer Science, volume 4894, pages 161--174, Springer, 2008
  
  • download: PDF (202.2 KB)
  • BibTeX entry: BibTeX entry
  • related info pages: Concept Learning in Description Logics
  • Best Student Paper Award ($1500 sponsored by the Machine Learning Journal, shared with the paper above)
  
  
• "Hybrid Learning of Ontology Classes"
  Jens Lehmann
  In: Proceedings of the 5th International Conference on Machine Learning and Data Mining (MLDM), Lecture Notes in Computer Science, volume 4571, pages 883-898, Springer, 2007
  
  • download: PDF (233.4 KB)
  • BibTeX entry: BibTeX entry
  • related info pages: Concept Learning in Description Logics
  
  

Theses

• "Comparison of Concept Learning Algorithms"
  Sebastian Hellmann
  University of Leipzig, 2008
  
  • download: PDF (865 KB)
  • BibTeX entry: BibTeX entry
  • related info pages: Concept Learning in Description Logics
  • related weblinks: DBpedia Homepage, DL-Learner homepage
  
  
• "Concept Learning in Description Logics"
  Jens Lehmann
  Diploma Thesis in Computer Science, TU Dresden, 2006
  
  • download: PDF (577.7 KB), PS (1.5 MB)
  • BibTeX entry: BibTeX entry
  • related info pages: Concept Learning in Description Logics
  • supervisors: Dr. habil. Pascal Hitzler, Prof. Steffen Hölldobler
  
  

Technical Reports

• "Ideal Downward Refinement in the EL Description Logic"
  Jens Lehmann, Christoph Haase
  Universität of Leipzig, 2009
  
  • download: PDF (238.9 KB)
  • BibTeX entry: BibTeX entry
  • related info pages: Concept Learning in Description Logics
  • related weblinks: DL-Learner homepage
  
  
• " A Refinement Operator Based Learning Algorithm for the ALC Description Logic"
  Jens Lehmann, Pascal Hitzler
  Universität Leipzig, 2007
  
  • download: PDF (257 KB)
  • BibTeX entry: BibTeX entry
  • related info pages: Concept Learning in Description Logics
  • An extended version including proofs of the ILP 2007 article with the same title.
  
  
• "Foundations of Refinement Operators for Description Logics"
  Jens Lehmann, Pascal Hitzler
  Universität Leipzig, 2007
  
  • download: PDF (233.5 KB)
  • BibTeX entry: BibTeX entry
  • related info pages: Concept Learning in Description Logics
  • A slightly older, but extended version including proofs of the ILP 2007 article with the same name.
  
  

Presentations

• "Ideal Downward Refinement in the EL Description Logic"
  Jens Lehmann, Christoph Haase
  19th International Conference on Inductive Logic Programming, 2009
  
  • download: PDF (767.9 KB)
  • related info pages: Concept Learning in Description Logics
  • related weblinks: DL-Learner homepage, ILP 2009
  
  
• "Presentation of AKSW Group and OWL Class Expressions Learning"
  Jens Lehmann
  Invited Talk at AIFB Karlsruhe, 2009
  
  • download: ODP (3 MB)
  • related info pages: Concept Learning in Description Logics
  • related weblinks: DL-Learner homepage
  
  
• "Hybrid Learning of Ontology Classes"
  Jens Lehmann
  5th International Conference on Machine Learning and Data Mining, 2007
  
  • download: PDF (382.3 KB)
  • related info pages: Concept Learning in Description Logics
  • related weblinks: MLDM 2007
  
  
• "ALC Concept Learning with Refinement Operators"
  Jens Lehmann, Pascal Hitzler
  17th International Conference on Inductive Logic Programming, 2007
  
  • download: PDF (1.2 MB)
  • related info pages: Concept Learning in Description Logics
  • related weblinks: ILP 2007
  
  
• "ALC Concept Learning with Refinement Operators"
  Jens Lehmann
  defense of diploma thesis, 2006
  
  • download: PDF (344.4 KB)
  • related info pages: Concept Learning in Description Logics
  
  
• "Lernen von ALC-Konzepten in Beschreibungslogiken und Ontologien"
  Jens Lehmann
  University of Leipzig, 2006
  
  • download: PDF (529.4 KB)
  • related info pages: Concept Learning in Description Logics
  
  
• "Genetic Programming and its use for Concept Learning in Description Logics"
  Jens Lehmann
  TU Dresden, 2006
  
  • download: PDF (494 KB)
  • related info pages: Concept Learning in Description Logics
  
  

Posters

• "Concept Learning in the ALC Description Logic"
  Jens Lehmann, Pascal Hitzler
  24th International Conference on Machine Learning, 2007
  
  • download: PDF (219.2 KB)
  • related info pages: Concept Learning in Description Logics
  • related weblinks: ICML 2007