Back to search

National Technical University of Athens (NTUA)

Σημασιολογικός εμπλουτισμός της πληροφορίας σε σχεσιακές βάσεις δεδομένων και συστήματα με επίγνωση περιβάλλοντος

Abstract

dc:description

The major problem of information available is that usually, the form in which it is published lacks both syntactical and semantic homogenisation. There is a huge amount of information, both online and in less interlinked sources. The existence of metadata that would add a higher degree of clarity is not taken for granted. The question that arises is how to integrate this volume information in order to produce a Knowledge Base from which assumptions could be derived, both explicit and implicit. The existing information management, processing and exploitation technologies are usually based on syntactic annotation. The current forms of annotation in their majority are monolithic and specialized approaches -ad hoc solutions- and offer limited potential of exporting results and conclusions. On one hand, they serve the purpose for which they were produced, but on the other hand they do not allow further exploitation in semantic level. An example is the Deep Web [1], which aggregates an enormous amount of information that is difficult to become useful and usable by the end user of the Web [2], Accordingly, the nature of the problem is similar in concept to the context-aware systems in which the volume ofinformation is growing at a great pace, hindering its management, integration with other sources ofinformation and hence the intelligent exploitation. This thesis focuses on the research in the broader context of semantic information enrichment. In particular, the research focuses on the changes and additions that need to be made at the technical and theoretical level, so that the next generation of information systems will integrate and exploit semantic information. As a solution towards the integration of information through semantic enrichment, the Semantic Web is adopted, which allows the addition of metadata which are subject to common standards and permit knowledge extraction from the existing one, and the combination of existing information in order to infer implicit knowledge, as well. The current work analyzes the problem of semantic information enrichment by presenting the components that compose it and add to the difficulty of solving it. In particular, two cases of the same problem are discussed in more detail. The first case concerns Database integration aiming at creating a distributed Knowledge Base that will incorporate intelligent features. The problem is not new, but it is considered in the light of recent technological developments and particularly in the expanding Semantic Web. The contribution of the thesis here lies in the analysis of the domain of Database and ontology collaboration, the detailed recording and categorizing of the approaches that exist in the bibliography and together they form the state of the art of the developments, and the analytical assessing of the conclusions drawn by close-examining this domain. It is noted that due to the dynamic developments in the domain, the recording of the approaches that exist in the bibliography contributes to the research in the area as a lack of relevant studies can be observed. In addition, an original approach to solving the problem is implemented and presented, and the limitations and possibilities of the systems facing the problem of Database and ontology collaboration are analyzed. The second case concerns context-aware systems. The substantial difference here in comparison with the Database integration lies in the time factor. Typically, these systems collect huge amounts of information that originates from data collected by sensors. While a low-level processing which may serve a limited number of requirements is easy, the problem also lies in how this information will be integrated and used in semantic level. For the study of the problem, a pilot middleware application is developed that is designed to process and semantically enrich information. The data which is used for the experiments conducted are semistructured multimedia metadata that come from heterogeneous distributed sources. By using the middleware, these data flow in the system forming a semantic Knowledge Base that is able to provide answers to queries at a higher semantic level. Specific reference is made to the real-time processing of the results, but also to the post procedures carried out in order to improve the system’s scalability. Also, the advantages and limitations resulting from the ensuing approach to creating a distributed Knowledge Base are analyzed and documented.

Degree

thesis:*
Grantor dc:publisher
National Technical University of Athens (NTUA)
Year dc:date
2009

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Konstantinou, Nikolaos
  • Κωνσταντίνου, Νικόλαος

Subjects

dc:subject × 18

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/18835

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Konstantinou, Nikolaos; Κωνσταντίνου, Νικόλαος. Σημασιολογικός εμπλουτισμός της πληροφορίας σε σχεσιακές βάσεις δεδομένων και συστήματα με επίγνωση περιβάλλοντος. National Technical University of Athens (NTUA), 2009. http://hdl.handle.net/10442/hedi/18835