Nwe Ni Tun

Semantic Enrichment in Ontologies for Matching

Matching between heterogeneous ontologies becomes crucial for interoperability in distributed and intelligent environments. Although many efforts in ontology mapping have already been conducted, most of them rely heavily on the meaning of entity names, rather than the semantics defined in ontologies. In order to deal with semantic heterogeneity, we propose a semantically enriched model of ontologies (called MetaOntoModel) where every domain concept is treated as a sort—an entity type that carries identity criteria for its instances—and classified these concepts based on three philosophical notions: identity, rigidity, and dependency. According to the classification, concept-level properties (called meta-knowledge) are embedded for each concept. Our novel idea is that if two concepts are semantically equivalent, then they have the same meta-knowledge. On the contrary, if two concepts possess different kinds of meta knowledge, then they cannot be matched. We also prove that meta-knowledge can determine not only the scope of semantic correspondences, but also which properties are the most relevant in finding correspondence between two similar concepts.

EnOntoModel: A Semantically-Enriched Model for Ontologies

Ontologies are intended to facilitate semantic interoperability among distributed and intelligent information systems where diverse software components, computing devices, knowledge, and data, are involved. Since a single global ontology is no longer sufficient to support a variety of tasks performed on differently conceptualized knowledge, ontologies have proliferated in multiple forms of heterogeneity even for the same domain, and such ontologies are called heterogeneous ontologies. For interoperating among information systems through heterogeneous ontologies, an important step in handling semantic heterogeneity should be the attempt to enrich (and clarify) the semantics of concepts in ontologies. In this article, a conceptual model (called EnOntoModel) of semantically-enriched ontologies is proposed by applying three philosophical notions: identity, rigidity, and dependency. As for the advantages of EnOntoModel, the conceptual analysis of enriched ontologies and efficient matching between them are presented.

Inheritance of multiple identity conditions in order-sorted logic

Guarino and Welty have developed the notion of identity condition (IC) as a subsumption constraint of ontological analysis, that is, IC must be inherited or carried among sorts In practical cases, a sort is often regarded to carry more than one identity condition via inheritance Thus, we provided the idea of multiple ICs in [8] Here, we extend our idea in order-sorted logic because this logic is one of the most promising ways to treat a sort ontology For this purpose, we reconsider the definition of identity and rigidity in terms of order-sorted language and possible world semantics Then, we propose an inheritance mechanism of identity conditions in order to solve subsumption inconsistency among sorts We present a practical example of knowledge management to illustrate the advantages of our formalism.

Identity conditions for ontological analysis

The role of ontologies is to provide a well-defined structure of domain knowledge that acts as the heart of any system of knowledge representation on that domain for the purposes of reasoning, knowledge sharing, and integration. Thus, it is essential to clarify the structure of knowledge in ontologies. In this paper, we discuss how ontology developers can define the identity conditions of classes explicitly, and can utilize them to develop structured taxonomies with adequate consistency. The background of this paper is OntoClean which is a domain independent methodology for ontology modeling using some philosophical notions. We exemplify the classification of sorts with necessary conceptual constraints. Then, we provide an explicit, simplified, and practical ontological analysis system regarding our subsumption constraints.

Consistency maintenance in ontological knowledge updating

In the formalization of ontology, the notion of identity is especially important, because only it can be explained by the set-theoretical model, in which each property is expressed as a set of individuals. Primarily, an identity condition (IC) is used to judge if two individuals are identical or not. In this paper, we consider multiple identity conditions for a sortal property in order to solve subsumption inconsistency, that is, lack of relevant ICs between two sortal properties. We employ modal logic and provide functions for identity conditions. Our theory of multiple ICs would contribute to the consistent updating of ontological knowledge not only by removing inadequate links on incompatible ICs but also by verifying the IC set of each sortal property on change of subsumption. We present a practical example to illustrate knowledge updating mechanism concerning with rigid properties and their ICs.

A philosophy‐driven entity classification and enrichment for ontology mapping

: Ontologies are intended to facilitate semantic interoperability among distributed and intelligent information systems. Because of the distributed nature of the World Wide Web, Web ontologies have been developing in multiple forms of heterogeneity. For interoperating among information systems through heterogeneous ontologies, ontology mapping is a prerequisite process to generate alignment between two ontologies. In order to improve alignment accuracy, our approach is to clarify and enrich the semantics of ontological entities before mapping. For this purpose, we present a semi-automatic framework of entity classification and enrichment by applying three philosophical notions: identity condition, existential rigidity, and external dependency. Our objective is to supply a set of philosophy-driven anchors into ontologies for their mapping process by using a sortal taxonomy as a background knowledge model.