Carlo Wouters

Semantic Completeness in Sub-ontology Extraction Using Distributed Methods

The use of ontologies lies at the very heart of the newly emerging era of Semantic Web. They provide a shared conceptual- ization of some domain that may be communicated between people and application systems. A common problem with web ontologies is that they tend to grow large in scale and complexity as a result of ever increasing information requirements. The resulting ontologies are too large to be used in their entirety by one application. Our previous work, M aterialized Ontology V iew E xtractor (MOVE), has addressed this problem by proposing a distributed architecture for the extraction/optimization of a sub-ontology from a large scale base ontology. The extraction process consists of a number of independent optimization schemes that cover various aspects of the optimization process. In this paper, we extend MOVE with a Semantic Complete- ness Optimization Scheme (SCOS), which addresses the issue of the semantic correctness of the resulting sub-ontology. Moreover, we utilize distributed methods to implement SCOS in a cluster environment. Here, a distributed memory architecture serves two purposes: (a). Facilitates the utilization of a cluster environment typical in business organizations, which is in line with our envisaged application of the proposed system and (b). Enhances the performance of the computationally extensive extraction process when dealing with massively sized realistic ontologies.

A distributed approach to sub-ontology extraction

The new era of semantic Web has enabled users to extract semantically relevant data from the Web. The backbone of the semantic Web is a shared uniform structure which defines how Web information is split up regardless of the implementation language or the syntax used to represent the data. This structure is known as an ontology. As information on the Web increases significantly in size, Web ontologies also tend to grow bigger, to such an extent that they become too large to be used in their entirety by any single application. This has stimulated our work in the area of sub-ontology extraction where each user may extract optimized sub-ontologies from an existing base ontology. Sub-ontologies are valid independent ontologies, known as materialized ontologies, that are specifically extracted to meet certain needs. Because of the size of the original ontology, the process of repeatedly iterating the millions of nodes and relationships to form an optimized sub-ontology can be very extensive. Therefore we have identified the need for a distributed approach to the extraction process. As ontologies are currently widely used, our proposed approach for distributed ontology extraction will play an important role in improving the efficiency of information retrieval.

MOVE: A Distributed Framework for Materialized Ontology View Extraction

AbstractThe use of ontologies lies at the very heart of the newly emerging era of semantic web. Ontologies provide a shared conceptualization of some domain that may be communicated between people and application systems. As information on the web increases significantly in size, web ontologies also tend to grow bigger, to such an extent that they become too large to be used in their entirety by any single application. Moreover, because of the size of the original ontology, the process of repeatedly iterating the millions of nodes and relationships to form an optimized sub-ontology becomes very computationally extensive. Therefore, it is imperative that parallel and distributed computing techniques be utilized to implement the extraction process. These problems have stimulated our work in the area of sub-ontology extraction where each user may extract optimized sub-ontologies from an existing base ontology. The extraction process consists of a number of independent optimization schemes that cover various aspects of the optimization process, such as ensuring consistency of the user-specified requirements for the sub-ontology, ensuring semantic completeness of the sub-ontology, etc. Sub-ontologies are valid independent ontologies, known as materialized ontologies, that are specifically extracted to meet certain needs. Our proposed and implemented framework for the extraction process, referred to as Materialized Ontology View Extractor (MOVE), has addressed this problem by proposing a distributed architecture for the extraction/optimization of a sub-ontology from a large-scale base ontology. We utilize coarse-grained data-level parallelism inherent in the problem domain. Such an architecture serves two purposes: (a) facilitates the utilization of a cluster environment typical in business organizations, which is in line with our envisaged application of the proposed system, and (b) enhances the performance of the computationally extensive extraction process when dealing with massively sized realistic ontologies. As ontologies are currently widely used, our proposed approach for distributed ontology extraction will play an important role in improving the efficiency of ontology-based information retrieval.

Ontology driven semantic profiling and retrieval in medical information systems

We propose the application of a novel sub-ontology extraction methodology for achieving interoperability and improving the semantic validity of information retrieval in the medical information systems (MIS) domain. The system offers advanced profiling of a users field of specialization by exploiting the concept of sub-ontology extraction, i.e., each sub-ontology may subsequently represent a particular user profile. Semantic profiling of a users field of specialization or interest is necessary functionality in any medical domain information retrieval system; this is because the (structural and semantic) extent of information sources is massive and individual users are only likely to be interested in specific parts of the overall knowledge documents on the basis of their area of specialization. The prototypical system, OntoMOVE, has been specifically designed for application in the medical information systems domain. OntoMOVE utilizes semantic web standards like RDF(S) and OWL in addition to medical domain standards and vocabularies encompassed by the UMLS knowledge sources.

A Practical Walkthrough of the Ontology Derivation Rules

To ensure the success of the semantic web, structures that are implemented, usually ontologies, should maintain their validity well into the future. These ontologies can be very extensive, introducing the problem that only parts (sub-ontologies) of them are needed by applications. A number of rules were established to get from such a base ontology to a derived sub-ontology. This paper explores these rules in a practical way, by giving a real-world scenario, and following it through from initial state to target ontology design. Each step and rule is applied to the scenario, and so the more practical side of the theoretical rules and steps is shown.

Large scale ontology visualisation using ontology extraction

The semantic web relies on ontologies to provide its required taxonomies. Often, ontologies tend to grow very large, introducing a number of problems. One of these problems is the difficulty in viewing and browsing of these ontologies by humans. Although visualisation techniques attempt to improve this by offering better graphical representations, they fail to fully resolve the issue, as they do not fully utilise the semantics that the ontology harbours. Ontologies are typically treated as graphs, which are unable to express and utilise several features that enable the rich semantics of the ontologies. This paper presents how these shortcomings can be overcome by reusing database principles. Solutions in the database for analogue problems such as Data Warehousing, to resolve information overload, are based on the notion of a view. This paper reinterprets this notion for ontologies, resulting in an ontology extraction methodology. This methodology uses optimisation schemes to allow integration and interpretation of semantic related features of ontologies. Using the methodology as a preprocessing step to visualisation, allows for better results for viewing and browsing large scale ontologies. A number of possible outcomes using this methodology are discussed.

A Service Oriented Architecture for Extracting and Extending Sub-Ontologies in the Semantic Grid

This paper presents a service oriented architecture (SOA) approach to a distributed framework for reusing, extracting and extending (tailoring) large domain ontologies in the semantic grid environment. The conceptual level of the framework describes how sub-ontologies are tailored while the architectural level of the framework describes the components of the framework that allows the tailoring process happen in the semantic grid environment. A prototype of the framework and a complexity evaluation measure are also provided.

Modeling ontology views: An abstract view model for semantic web

The emergence of Semantic Web (SW) and the related technologies promise to make the web a meaningful experience. However, high level modeling, design and querying techniques proves to be a challenging task for organizations that are hoping to utilize the SW paradigm for their industrial applications. To address one such issue, in this paper, we propose an abstract view model with conceptual extensions for the SW. First we outline the view model, its properties and some modeling issues with the help of an industrial case study example. Then, we provide some discussions on constructing such views (at the conceptual level) using a set of operators. Later we provide a brief discussion on how such this view model can utilized in the MOVE [1] system, to design and construct materialized Ontology views to support Ontology extraction.