National Technical University of Athens (NTUA)
Παραγωγή σημασιολογικού περιεχομένου από δομημένες και ημιδομημένες πηγές δεδομένων
Abstract
dc:descriptionThe World Wide Web has become a part of everyday life, having changed the way people communicate, as well as the way we create, share and search for information. The Semantic Web seeks to revolutionize the way this huge amount of available information is used, allowing automated procedures to make use of its meaning, infer new facts and integrate it with other information in a way that is meaningful to the end user. However, the Semantic Web vision has not been fully materialized yet and one of many possible reasons is the lack of an adequate critical mass of data that can be readily used by semantic applications. Therefore, this thesis investigates the issue of semantic content generation from existing structured and semistructured data sources, with the ultimate goal of bringing this entire wealth of information in the Semantic Web. First of all, an extended literature survey is carried out in order to clarify the various aspects of relational database usage in the Semantic Web context and identify all relevant issues, such as ontology generation from a relational schema, the export of relational database contents in the form of an RDF graph and the discovery of mappings between a relational database and an ontology. A simplified mapping system that merely uses SQL queries for the mapping definition is then proposed and the theoretical and practical shortcomings of this approach are pointed out. Furthermore, an algorithm for the rewriting of SPARQL queries to semantically equivalent SQL ones in the presence of an R2RML mapping is analysed, allowing for dynamic access of relational database contents via semantic queries. Finally, the problem of semantic annotation and processing of semistructured data from dynamic sources is investigated and an appropriate extension to an already implemented context-aware system is proposed. This extension applies windowing techniques in the incoming data stream in order to keep the response time of the system under acceptable levels.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2013
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Spanos, Dimitrios-Emmanuel
- Σπανός, Δημήτριος-Εμμανουήλ
Subjects
dc:subject × 27- Σημασιολογικός ιστός
- Σχεσιακές βάσεις δεδομένων
- Οντολογίες
- Αντιστοιχία
- Εικονική βάση γνώσης
- Ροή δεδομένων
- Δίκτυα αισθητήρων
- Παράθυρο οντολογικών ατόμων
- Semantic web
- Relational databases
- Ontologies
- Resource Description Framework (RDF)
- Mapping
- Virtual knowledge base
- SPARQL
- R2RML
- Data stream
- Sensor networks
- Individual window
- Φυσικές Επιστήμες
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Natural Sciences
- Computer and Information Sciences
- Engineering and Technology
- Electrical Engineering, Electronic Engineering, Information Engineering
Rights
dc:rights- Statement dc:rights
-
- BY_NC_SA
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/29105