Aristotle University Of Thessaloniki (AUTH)
Τεχνικές υποστήριξης της εξατομίκευσης στο διαδίκτυο
Abstract
dc:descriptionThe tremendous growth of the World Wide Web has generated an enormous amount of information, which cannot be consumed by the users. This phenomenon has been described as Information Overload. Due to this, the internet users face difficulties in finding the information they need and assess its reliability. Adaptation of the available information according to user needs and interests can help users effectively manage the volume of information and assess its reliability. Hence, they can make decisions that maximize their satisfaction. In order to tackle the aforementioned challenges, the current thesis has investigated techniques for web personalization. In particular, we present a recommender system for mass customization, which suggests a product configuration based on user affective needs that comprise customers’ needs for a product driven by their emotions. The technique is built on top of multiple classifiers, one per product design parameter, leading in an increased overall accuracy, thus allowing for its application also for products that were not included in the original training data set. Moreover, we investigated techniques for supporting the personalized management of Web Services. Specifically, we presented a technique for Semantic Classification of web services based on the domain to which they belong, but also for classification of the elements that a web service includes into ontology concepts. Our goal was to support Web Service discovery based on users’ needs by the means of the proposed technique. After empirical experimentation we discovered that our technique achieved high classification accuracy compared to similar existing approaches. For supporting the provision of Web Services in the Cloud, an inference engine was implemented for reliability assessment of services, service providers and consumers, based on the combination of statistical data regarding the use of Cloud services, with user feedback. Finally, we present a technique for recommending personalized discount schemas based on reliability and loyalty assessment of both providers and consumers.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2017
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Mavridou, Efthimia
- Μαυρίδου, Ευθυμία
Subjects
dc:subject × 12- Εξατομίκευση διαδικτύου
- Υπηρεσίες ιστού
- Εξόρυξη γνώσης
- Σημασιολογική ταξινόμηση
- Web personalization
- Web services
- Data mining
- Semantic classification
- 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/41981