Aristotle University Of Thessaloniki (AUTH)
Συλλογή και ανάλυση δεδομένων από διαδικτυακά κοινωνικά μέσα με χρήση σημασιολογικών μεθόδων
Abstract
dc:descriptionThis PhD thesis focuses on the areas of Semantic Web and Social Media Analysis and aspires to enhance their respective advantages. The main goal of this thesis is to contribute towards the realization of Semantic Web by introducing semantic methods for the collection, analysis, and management of semantic content. The concept of semantic analysis covers both the analysis using natural language processing techniques, as well as the analysis with semantic web methods. These two approaches can be applied in any order to make data understandable by both machines and humans. The thesis’ objectives include a) the definition of a methodology and the implementation of an adaptable system for collecting data from web social media, b) the definition of a methodology for the semantic process of web social information for the identification of events, and c) the definition of a methodology and the development of a semantic system that integrates data stored in traditional formats with data collected from the Web and data already available in semantically aware formats. Based on these goals, our thesis focuses on the development of a framework, accompanied with a set of algorithms that address each of the above mentioned problems. Thus, we present a web crawler for collecting data using semantically aware techniques, a methodology for the semantic analysis of data to identify all the real world events described by users in online social media, and a system for the integration of semantic and legacy data stored in relational databases. In conclusion, our contribution includes the following: first, we propose a methodology for the identification of events enriched with semantic information using data collected from the Social Web; second, we design and implement an innovative and efficient event detection algorithm that is applicable to large data streams; third, we propose a methodology to efficiently identify events of a predefined topic, which was evaluated in an international challenge and outperformed other approaches; fourth, we design and implement an integrated framework for the management of semantic data including integrating, modifying, searching, viewing, editing and inferencing on semantic data, and fifth, we design and implement a mechanism for the collection of semantic and legacy data from the web, which can adapt to the evolutionary nature of the WWW using the semantic information scattered on it.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2013
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Vavliakis, Konstantinos
- Βαβλιάκης, Κωνσταντίνος
Subjects
dc:subject × 14- Σημασιολογικός ιστός
- Κοινωνικός ιστός
- Εξόρυξη δεδομένων
- Σχεσιακές βάσεις δεδομένων
- Αναγνώριση γεγονότων
- Semantic web
- Social web
- Data mining
- Relational databases
- Event detection
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Engineering and Technology
- Electrical Engineering, Electronic Engineering, Information Engineering
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/37939