University of Western Macedonia
Δίκτυα επικοινωνιών και υπηρεσίες με επίγνωση κοινωνικού περιβάλλοντος χρήστη
Abstract
dc:descriptionThe constant evolution of social networks offers high potential in using them for the purpose of designing and offering advanced services not just to the end user but to the whole technological infrastructure. The specific services can adjust their decisions and functionalities based on the natural, social, and computing environment that are exposed to, which satisfies with the best way the end user requirements and preferences, while they are improving the efficiency of the technological infrastructure. In this thesis, two research interests are studied a) how the social information contributes to the development of social network applications, specifically for recommendation systems and b) how we can use the social information to improve the performance of network services and operations, by emphasizing on routing protocols and vehicular networks.During the last decade, there are two types of social networks that have become very popular: a) the online networks, where users, get connected, create relationships, share content, exchange opinions and experiences and b) the offline social networks, that exist in the real physical world, which are using occasional contacts and connections without proper infrastructure between devices (e.g., mobile phones, vehicles). Has been proven that people that are socially closer tend to interact more often and for a longer period in comparison with other users.Advanced applications, that combine geographical data, mobile information, social relationships, and information deriving from user profiles are being suggested, with main goal to provide service delivery with improved quality. In this thesis, recommendation systems are proposed, where based on the social relationships of the users, that are relying on trust and have an impact on a social network, providing customized rankings and recommendations for each user. Both the positive and negative relationships of the users are integrated, by considering the time dimension, while the problem of the cold start of the recommendation systems is efficiently addressed. Moreover, in contrast with previous studies that had been using just the global influence coming from the users to provide appropriate recommendations, in this thesis the user’s local influence is introduced as a metric for the proposed ranking algorithm. It is proved that the combination of the social information with the local influence, contributes to the ability to provide higher quality customized recommendations. The social information is being capitalized to design network services using VANET (Vehicular Ad-Hoc NETworks – VANET). Considering the characteristics of the VANET networks along with the possible selfish or/and malicious behaviour of the nodes on this kind of networks, metrics are suggested that would help to calculate the social proximity between users both on the online and the physical world. Additionally, taking into account the social metrics mentioned above, efficient routing protocols are proposed. Concluding, incorporating the social information in the design of routing protocols improves the performance of the routing in this type of networks, and at the same moment overcomes cases where nodes might behave selfish or/and malicious.
Degree
thesis:*- Grantor dc:publisher
- University of Western Macedonia
- Year dc:date
- 2022
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Mantas, Nikolaos
- Μαντάς, Νικόλαος
Subjects
dc:subject × 8- Οχηματικά δίκτυα
- VANETs
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Επιστήμες Μηχανικού και Τεχνολογία
- Μηχανική και συστήματα επικοινωνιών, Τηλεπικοινωνίες
- Electrical Engineering, Electronic Engineering, Information Engineering
- Engineering and Technology
- Communication engineering and systems, Telecommunications
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/52580