Syntactic regularities are repetitive structures of axioms in the asserted form of an ontology. In this paper, we present a Prote´ ge´ 4 plugin based on the RIO (Regularities Inspector for Ontologies) framework, which detects such regularities in ontologies using cluster analysis. The RIO plugin is an open source tool for visualising syntactic regularities in ontologies; the user can load them in Prote´ ge´ 4 and view clusters of similar entities, generalisations describing the clusters and abstracting axioms. The RIO plugin can be used for gaining an intuition about the embedded patterns in ontologies and deviations from these patterns. Availability: the RIO plugin is available for download from http: //riotool.sourceforge.net/.
DESCRIPTION A syntactic regularity is defined as a set of axioms with reoccurring (regular) syntactic structure. A regularity can be expressed with a generalisation, which is an axiom that allows for variables to replace entities. For example, given the following axioms from the AminoAcid ontology1:
Then the syntactic regularity of these axioms can be given by the following generalisation:
?AminoAcid SubClassOf hasSize some ?Size where ?AminoAcid is a variable holding the amino acids in the axioms fA, I, E g and ?Size is a variable holding the corresponding sizes fTiny, Large, Small g. Such a framework can be used when authoring an ontology, in order to pinpoint repetitive information. RIO was presented in Mikroyannidi et al. (2011) where further detail may be found about RIO and its application to SNOMED.
In this paper, we present the RIO plugin; a Prote´ge´-4 plugin for inspecting syntactic regularities in ontologies. It can be useful for revealing parts of an ontology which have been developed in the same way. Moreover, it can be useful for gaining an intuition about the construction of the ontology. mikroyannidijiannonejstevens@cs.manchester.ac.uk 1 http://www.co-ode.org/ontologies/amino-acid/ 2006/05/18/amino-acid.owl
The RIO framework uses cluster analysis for the detection of syntactic regularities. The purpose of cluster analysis is to partition data into groups (clusters) that are meaningful, useful, or both Tan et al. (2005). Thus, the computation of the clusters is based on the similar usage of entities in axioms. Finally, generalised axioms express the syntactic regularities, which are a synthetic view of all the axioms that contribute to the generation of an entity cluster.
The RIO plugin is an open source project. Its implementation is based on OWLAPI2. A standalone Java tool is used for the regularities’ computation, which are then saved in an XML file. The user can load and visualise the regularities in Prote´ge´-4 using the RIO plugin.
Figure 1 shows one of the main views of the RIO plugin in Prote´ge´-4. The cluster view shows the name of every cluster and the number of entities in the cluster. The Cluster member list view shows the list of entities in a cluster. The generalisation view shows information about the regularities that are captured for every cluster. In the generalisation view, every generalisation can be unfolded to show the axioms that are abstracted (instantiations) and metrics about the variables. The metrics that are shown are the number of instantiations, and the number of entities that are covered by each variable. In principle, the union of the generalisations in the view describes a cluster, but a single generalisation is not necessarily applicable to all members of a cluster.
The recognition of syntactic regularities should be helpful in understanding the composition of an ontology, as it can reveal parts of the ontology that were designed in similar ways. This should help the user to complete tasks, such as extensions of the ontology, its integration with other ontologies, quality assurance and so on. 2 http://owlapi.sourceforge.net