National Technical University of Athens (NTUA)
Βελτιστοποίηση συστημάτων εκπαίδευσης και συλλογικής ευφυΐας βασισμένη σε μοντελοποίηση συμπεριφοράς και τεχνικές μηχανικής μάθησης
Abstract
dc:descriptionThe present doctoral thesis focuses on the scientific fields of e-learning and collaborative collective intelligence systems and presents novel methods for improving these systems. These methods are established through the use of machine learning techniques and through behaviour modelling of online users. As far as e-learning is concerned, the thesis presents the limitations that instructors face in promptly and accurately predicting the future behaviour of trainees, mainly due to the distance that is involved in this type of education. This inability to predict trainee behaviour results in difficulties when addressing the specialized educational needs of the trainees, which in turn leads to increased dropout rates, compared to traditional education. To address the above problem, two novel methods are established in the present thesis, for early and dynamic prediction of the final achievement of e-learning students and their possibility to drop the online course. The methods are based on the use of machine learning techniques and the modelling of the trainee behaviour, aiming to provide instructors with the means to address the specialized educational needs of the trainees in a prompt and accurate manner, thus contributing to die increase of e-learning retention. Apart from the problem of reduced feedback due to the distance involved, one of the most important issues of e-learning and of Web systems in general, is the quality assurance of the information communicated to the users. This problem is particularly observed in collaborative collective intelligence systems, and more specifically in Web 2.0 systems, which assist the cooperation of large numbers of users on the task of content creation. Although these systems allow participating users to collaborate, they do not succeed in co-ordinating them and in taking full advantage of their knowledge and skills, so that content of the best possible quality will emerge out of their collective intelligence. To address the aforementioned issue, the present thesis proposes a new self-regulating collective intelligence system that acts towards ensuring content quality, by combining machine learning techniques with the collective intelligence of its peer users.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2009
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Lykourentzou, Ioanna
- Λυκουρέντζου, Ιωάννα
Subjects
dc:subject × 18- Τηλεκπαίδευση
- Μηχανική μάθηση
- Συλλογική ευφυΐα
- Συνεργατικά συστήματα
- Συστήματα ιστού δεύτερης γενιάς
- E-learning
- Machine learning
- Collective intelligence
- Collaborative systems
- Web 2.0
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Κοινωνικές Επιστήμες
- Εκπαίδευση
- Engineering and Technology
- Electrical Engineering, Electronic Engineering, Information Engineering
- Social Sciences
- Education
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/27170