Back to search

Aristotle University Of Thessaloniki (AUTH)

Ανάπτυξη τοποκεντρικών εφαρμογών ηλεκτρονικού εμπορίου με τη χρήση τεχνολογιών σημασιολογικού ιστού

Abstract

dc:description

Regarding this thesis, the mobile commerce applications’ area and its impact on the economy is studied. In addition, Semantic Web technologies and personalization techniques (that are constantly used in such applications) are taken into consideration. The purpose of this thesis was to discover and demonstrate all the related research issues to these sectors (e.g. economic and business models, impact on various sectors of the economy, personalized information, security, etc.), and, afterwards, to propose innovative methodologies (that will find practical application through the development of state-of-the-art systems) to alleviate these issues.After reviewing the above areas, two systems that provide personalized information (based on the use of Semantic Web technologies) are presented.The first system is called SPLIS (Semantic Personalized Location Information System) and its aim is to connect POI owners’ marketing policies with regular users so as to provide personalized offers to them. The main point of the methodology is to model marketing policies via logical rules. As soon as POI owners (via an innovative interface) place their offers, the system transforms these into machine understandable rules. Afterwards, the system executes these rules/offers and provides qualitative contextualized information to the users.The second system is called G-SPLIS (GeoSocial Semantic Personalized Location Information System). In this system, which has been built on top of SPLIS, is proposed (apart from the POI owners’ policies) to model regular users’ preferences with logical rules. In detail, regular users can insert their preferences via an innovative web-based editor. These preferences are transformed to logical rules and are combined with those of the owners. Additionally, social network functionalities have been implemented and are combined with human knowledge and Semantic Web technologies (e.g. sharing rules). The proposed methodology, which is demonstrated through the above systems, contributes to the area of mobile commerce application applications by alleviating the above issues: •Providing more qualitative personalized information. •Commercialization and the development of suitable business models. •Interoperability. •Designing and developing the knowledge base of a system. •Designing and implementing more efficient rules. Apart from the above, a novel hybrid approach (which combines association rules, probability metrics and user’s context) to alleviate the cold start problem in rule-based context aware recommender systems is proposed. The approach combines community generated knowledge (e.g. Users’ check-in history) with Semantic Web technologies and association rule mining algorithms. At first, association rule mining algorithms are used to extract rules from one user at a time, and not from the entire database of users’ check-ins. Semantic web technologies such as ontologies and rules are used to represent the entities that are involved in the rules. After that a novel scoring function is used to rank the rules and recommend the most suitable of them to each user. The presented approach exhibits the following advantages/contributions: •Alleviates the cold start problem. •Assists in providing more accurate context-aware suggestions.•Because it does not recommend specific points of interest directly, but patterns that are related to a set of points of interest, it addresses the issues of diversity and novelty (regarding the recommendations). •Helps in designing and implementing more efficient rules. •Contributes to the issue of interoperability between heterogeneous systems.

Degree

thesis:*
Grantor dc:publisher
Aristotle University Of Thessaloniki (AUTH)
Year dc:date
2018

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Viktoratos, Iosif
  • Βικτωράτος, Ιωσήφ

Subjects

dc:subject × 10

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/43740

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

Viktoratos, Iosif; Βικτωράτος, Ιωσήφ. Ανάπτυξη τοποκεντρικών εφαρμογών ηλεκτρονικού εμπορίου με τη χρήση τεχνολογιών σημασιολογικού ιστού. Aristotle University Of Thessaloniki (AUTH), 2018. http://hdl.handle.net/10442/hedi/43740