Upper ontology

About: Upper ontology is a research topic. Over the lifetime, 9767 publications have been published within this topic receiving 220721 citations. The topic is also known as: top-level ontology & foundation ontology.

A Text Categorization Perspective for Ontology Mapping

Abstract: This position paper addresses the problem of ontology mapping which is pervasive in context where semantic interoperability is needed. A preliminary solution is proposed using external information, i.e. documents assigned to the ontology to calculate similarities between concepts in two ontologies. Text categorization is used to automatic assign documents to the concepts in the ontology. Based on the similarities measure, a heuristic method is used to establish mapping assertions for the two ontologies. 1. Background & Problem Lately, there has been much research related to the new generation web – semantic web. The hope is that the semantic web can alleviate some of the problems with the current web, and let computers process the interchanged data in a more intelligent way. In an open system like the Internet, which is a network of heterogeneous and distributed information systems (IS), mechanisms have to be developed in order to enable systems to share information and cooperate. This is commonly referred to as the problem of interoperability. The essential requirement for the semantic web is interoperability of IS. If machines want to take advantage of the web resources, they must be able to access and use them. Ontology is a key factor for enabling interoperability in the semantic web [Bernees-Lee01]. An ontology is an explicit specification of a conceptualisation [Uschold96]. It includes an explicit description of the assumptions regarding both the domain structure and the terms used to describe the domain. Ontologies are central to the semantic web because they allow applications to agree on the terms that they use when communicating. Shared ontologies and ontology extension allow a certain degree of interoperability between IS in different organizations and domains. However there are often cases where there are multiple ways to model the same information and the problem of anomalies in interpreting similar models leads to a greater complexity of the semantic interoperability problem. In an open environment, ontologies are developed and maintained independently of each other in a distributed environment. Therefore two systems may use different ontologies to represent their view of the domain. Differences in ontologies are referred to as ontology mismatch [Klein01]. The problem of ontology mismatch arises because a universe of discourse, UoD, can be specified in many different ways, using different modelling formalisms. In such a situation, interoperability between systems is based on the reconciliation of their heterogeneous views. How to tackle ontology mismatch is still a question under intensive research. As pointed out in [Wache01], three basic architectures to cope with ontology mismatch can be identified: single ontology approaches, multiple ontologies approaches and hybrid approaches. An illustration of each of them is given in figure 1. Global ontology Local ontology Local ontology Local ontology Local ontology Local ontology Local ontology Global shared vocabulary Figure 1a: Single ontology approach Figure 1b: Multiple ontologies approach Figure 1c: Hybrid approach Figure 1 Architectures to cope with ontology mismatch. In the Single ontology approach, a global ontology provides a shared global ontology to specify the semantics. All systems or information sources are related to the global ontology i.e. they are unified. The global ontology can be a combination of modularised sub ontologies. In the Multiple ontologies approach, each information source has its own local ontology, which doesn’t necessarily use the same vocabulary. Each ontology can be developed independently because there is a loose coupling between the ontologies. To achieve interoperability the ontologies must be brought together by mapping rules (links). In the Hybrid approach, the basic features of the two other approaches are combined in order to overcome some of their disadvantages. Like the multiple ontologies approach, each source has its own local ontology. But the local ontologies are developed from a global shared vocabulary in order to make the alignment of ontologies easier. The shared vocabulary defines basic terms for the domain, which can be combined to describe more complex semantics in the local ontologies. A single ontology approach, which is based on tight coupling, most often is too rigid and does not scale well in a large open environment. Adding a new source will most often lead to a new unification process [Wiederhold99]. In our opinion, a multiple or hybrid approach is more appropriate, allowing a degree of local autonomy to coexist with partial interoperability. In both of the latter cases, developing means to facilitate mapping between two ontologies is necessary. A web portal scenario can be used to illustrate the ontology mapping problem. A Web portal is a web site that provides information content on a common topic, for example a specific city or a specific interest (like ski). A web portal allows individuals that are interested in the topic to receive news, find and talk to other interested people, build a community, and find links to web resources of common interest. Normally, web portals can define an ontology for the community. This ontology defines terminologies for describing content and serves as an index for content retrieval. One example of an ontology-based portal is The Open Directory Project [ODP], a large, comprehensive human-edited directory of the Web. Say, for example, that there are two web portals about topic sports. One of them uses ODP, while the other is based on a sub portion of Yahoo! Category. Users may want to share or exchange information between the portals. In that context, means that allow ontologies to map terms to their equivalents in other ontologies, must be developed.

A semantic web service-based architecture for the interoperability of e-government services

Abstract: We propose a semantically-enhanced architecture to address the issues of interoperability and service integration in e-government web information systems. An architecture for a life event portal based on Semantic Web Services (SWS) is described. The architecture includes loosely-coupled modules organized in three distinct layers: User Interaction, Middleware and Web Services. The Middleware provides the semantic infrastructure for ontologies and SWS. In particular a conceptual model for integrating domain knowledge (Life Event Ontology), application knowledge (E-government Ontology) and service description (Service Ontology) is defined. The model has been applied to a use case scenario in e-government and the results of a system prototype have been reported to demonstrate some relevant features of the proposed approach.

Towards an ontology modeling tool. A validation in software engineering scenarios

Abstract: Ontology creation and management related processes are very important to define and develop semantic services. Ontology Engineering is the research field that provides the mechanisms to manage the life cycle of the ontologies. However, the process of building ontologies can be tedious and sometimes exhaustive. OWL-VisMod is a tool designed for developing ontological engineering based on visual analytics conceptual modeling for OWL ontologies life cycle management, supporting both creation and understanding tasks. This paper is devoted to evaluate OWL-VisMod through a set of defined tasks. The same tasks also will be done with the most known tool in Ontology Engineering, Protege, in order to compare the obtained results and be able to know how is OWL-VisMod perceived for the expert users. The comparison shows that both tools have similar acceptation scores, but OWL-VisMod presents better feelings regarding user's perception tasks due to the visual analytics influence.

Building a comprehensive ontology to refine video concept detection

Abstract: Recent research has discovered that leveraging ontology is an effective way to facilitate semantic video concept detection. As an explicit knowledge representation, a formal ontology definition usually consists of a lexicon, properties, and relations. In this paper, we present a comprehensive representation scheme for video semantic ontology in which all the three components are well studied. Specifically, we leverage LSCOM to construct the concept lexicon, describe concept property as the weights of different modalities which are obtained manually or by data-driven approach, and model two types of concept relations (i.e., pairwise concept correlation and hierarchical relation). In contrast with most existing ontologies which are only focused on one or two components for domain-specific videos, the proposed ontology is more comprehensive and general. To validate the effectiveness of this ontology, we further apply it to video concept detection. The experiments on TRECVID 2005 corpus have demonstrated a superior performance compared to existing key approaches to video concept detection.

Characterizing Quality of Knowledge on Semantic Web

Abstract: The Semantic Web is intended for knowledge sharing among agents as well as humans. To achieve this goal, Ontologies, which express knowledge in a certain vitality as well as in a machine interpretable form, were introduced. The growing demand for facilitating deployment and reuse of Ontologies has increased the need to develop adequate criteria to measure the quality of Ontologies conceptualizing a domain. Our research is motivated by the urgent needs of rigorous mechanisms which analyze ontological features from diverse perspectives and determine their quality levels. This paper presents the methodology representing qualitative and quantitative analysis of Ontologies and their classification. The Ontological tools, which were implemented based on the methodology, are proposed to provide multiple interfaces to humans and agents, thus supporting Ontology Engineering process. The proposed framework has gone through a great deal of testing and evaluation processes in the context of a real application of Ontology analysis and classification.