Debugging ontologies and mappings in ontology networks

Developing ontologies is not an easy task and, as the ontologies grow in size, they are likely to show a number of defects. Such ontologies, although often useful, also lead to problems when used in semantically-enabled applications. Wrong conclusions may be derived or valid conclusions may be missed. Defects in ontologies can take different forms. Syntactic defects are usually easy to find and to resolve. Defects regarding style include such things as unintended redundancy. More interesting and severe defects are the modeling defects which require domain knowledge to detect and resolve such as defects in the structure, and semantic defects such as unsatisfiable concepts and inconsistent ontologies. Further, during the recent years more and more mappings between ontologies with overlapping information have been generated, e.g. using ontology alignment systems, thereby connecting the ontologies in ontology networks. This has led to a new opportunity to deal with defects as the mappings and other ontologies in the network may be used in the debugging of a particular ontology in the network. It also has introduced a new difficulty as the mappings may not always be correct and need to be debugged themselves.

In this tutorial we give an overview of current approaches for debugging semantic and structure defects in ontologies (single or in an ontology network) as well as approaches for debugging mappings.

Target audience

The tutorial is designed for researchers and practitioners dealing with ontology development, ontology mapping, ontology evolution, or ontology engineering in general.

A familiarity with OWL and description logics is an advantage, although we will introduce the necessary basics.

Content

The slides are available here.

1. Overview of defects in ontologies and ontology networks
We will introduce the different kinds of defects in ontologies with a focus on semantic defects and defects in the is-a structure of ontologies. We will exemplify these kinds of defects, and discuss the consequences of these defects in semantically-enabled applications.

2. Debugging semantic effects in ontologies
Generally, we can distinguish two kinds of logical contradictions in ontologies: inconsistency and incoherence. An ontology is inconsistent iff it has no model. An ontology is incoherent iff there exists some unsatisfiable concept. In this part of the tutorial we give an overview of the state of the art in dealing with these defects in ontologies. We will introduce methods for debugging a single inconsistent or incoherent ontology, introduce some optimizations techniques, and introduce methods that regard the debugging problem as a revision problem for description logics terminologies.

3. Debugging ontology mappings
The quality of an ontology mapping is mainly quantified in terms of precision and recall comparing the mapping against a gold standard. In this part of the tutorial we show how to debug an incoherent mapping. The techniques that we discuss do not only result in a coherent and usable mapping, but can also increase precision and recall of the resulting mapping. We show how to integrate a reasoning-based approach in an ontology matching system as a final layer and discuss the effects of different strategies at hand of a running example. In particular, we show how to use a state-of-the art reasoner to implement the required algorithms.

4. Debugging missing is-a structure
In addition to its importance for the correct modeling of a domain, the is-a structure in ontologies is also important in semantically-enabled applications, e.g. in ontology-based search and annotation. In this part of the tutorial we present a debugging approach that includes detection of missing is-a structure and repair of missing is-a structure. We will focus on algorithms for generating, recommending and executing repairing actions. Further, we introduce RepOSE, a semi-automatic system for debugging missing is-a structure in ontologies.

Organizers

Patrick Lambrix is a professor at Linköping University, Sweden. He is PI of the project that developed RepOSE, a system for debugging missing is-a structure in ontologies. To our knowledge, RepOSE is the first system that generates suggestions on how to repair missing is-a structure in a more advanced way than just adding the missing structure. He is also PI for the project that developed the ontology alignment system SAMBO, winner OAEI 2008 Anatomy Track. In addition to different alignment strategies, methods for recommendation of ontology alignment strategies and approaches for using partial reference alignments in the alignment process were developed.

Guilin Qi is a professor at Southeast University in China. He has participated in EU FP7 project NeOn: Lifecycle Support for Networked Ontologies and developed a tool for diagnosing and repairing ontology networks (RaDON). He received his PhD in Computer Science from Queen's University of Belfast in 2006. His research interests include knowledge representation and reasoning, and semantic web. He has published more than 60 papers in these areas, many of which published in proceedings of major conferences. He is in the editorial board of the Journal of Web Semantics and is an associate editor of the Journal of Advances in Artificial Intelligence. He has organized several international workshops and served as PC member of several international conferences and workshops, such as IJCAI'11, AAAI'11, ESWC'11, ISWC'09, ISWC'10. He has given a tutorial about finding justifications of entailments in OWL ontologies at ASWC'09 and a tutorial about large scalable ontology reasoning at AAAI'10.

Christian Meilicke is a research assistant at the Chair of Artificial Intelligence at the University of Mannheim, Germany. He graduated from the University of Mannheim as a Master in Philosophy and Educational Science in 2003. After moving from the field of humanities to in the field of computer science, he is working now for his PhD on the topic of reasoning-based debugging of ontology mapping. For several years he has been organizing together with Heiner Stuckenschmidt the anatomy track of the Ontology Alignment Evaluation Initiative (OAEI). Has has published several award-winning papers on the topic of this tutorial. His work is mainly concerned with the adoption of reasoning-based methods to the field of ontology matching.