| Author: | Miao, Xue |
| Title: | Multimedia ontology integration based-on RDF graph mapping |
| Degree: | M.Sc. |
| Year: | 2009 |
| Subject: | Hong Kong Polytechnic University -- Dissertations. Multimedia systems. Semantic Web. Metadata. Information storage and retrieval systems. |
| Department: | Department of Computing |
| Pages: | viii, 139 leaves : ill. ; 30 cm. |
| Language: | English |
| Abstract: | Nowadays, the more multimedia information becomes accessible, the harder it is to identify and find what you want, and the more difficult to operate information. The main reason is the problem of semantic heterogeneous. That is, the different sources information use a variety of terms to express the same concept; or the same term in different sources information express different meanings. Therefore, people choose to replace the previous text-based engines with ontology to express multimedia information. Since ontology has a strong ability to express semantics concept and to acquire knowledge. It not only accurately describes the meaning of concept but also describes the intrinsic correlation between concepts. Furthermore, using ontology to represente and annotate multimedia contents have been identified as an important step towards more efficient manipulation and retrieval multimedia. Therefore, using ontology to integrate multimedia information can better solve the problem of semantic heterogeneous. Although the large number of the researches and application of ontology matching gradually become a hot topic, it still has many challenges. The thesis focuses on improving the accuracy of automated multimedia ontology matching algorithm. We will use the RDF graphs to integrate the two multimedia ontologies by graph mapping. As the foundation of semantic Web is resource description framework (RDF), besides more and more information is described in RDF format. This is a very popular and practical choice for matching Web ontologies expressed by RDF(S) and OWL. It is possible to create RDF Graph by using RDF sentence with ontology markup languages such as OWL. In OWL, we combine the function of description multimedia information provided by MPEG-7 to define an OWL notes properties to describe these features, the values of these attributes is by calling on MPEG-7 describe instance characters. Ontology matching is the operation that constructs the mapping of two ontologies, and an efficient and rigorous similarity measure is a pre-requisite of an ontology matching process. Based on the characteristics of RDF and several restrictions on the RDF graph, a matching algorithm (OMR) is proposed for matching ontology. It used Resource Description Framework (RDF) graph to present ontology and converted the ontology matching problem to building RDF graph matching tree. Moreover, the proposed approach expressed a similarity measure based on the entities of nodes from the matching tree to obtain the mapping of the two ontologies. Implementation and experimental results were shown to demonstrate the effectiveness of the matching approach. OMR algorithm, a prominent component of the multimedia ontology matching, is an automated ontology matching system to help achieve interoperability between the Semantic Web applications using different but interrelated ontology. It uses RDF graphs to show ontologies and computes structural similarities between domain entities and between statements (triples) in ontologies by recursively propagating similarities in the bipartite graphs. OMR needs a set of external lists as input and step by step produces additional lists as output. Multimedia ontology matching is a Semantic Web infrastructure application, which is the management of the large-scale multimedia collections using OMR to replace the semantic metadata integration techniques with ontology-driven application of Semantic Web technology in the multimedia field. The main objective of this system is to provide the basic skills for finding, adjustment and learning the ontology, and ultimately capturing the multimedia information from the network through an ontology-driven approach. |
| Rights: | All rights reserved |
| Access: | restricted access |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| b23181783.pdf | For All Users (off-campus access for PolyU Staff & Students only) | 3.57 MB | Adobe PDF | View/Open |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/3933