Your browser does not support JavaScript!

Home    Search  

Results - Details

Search command : Author="Στεφανίδης"  And Author="Κωνσταντίνος"

Current Record: 8 of 72

Back to Results Previous page
Next page
Add to Basket
[Add to Basket]
Identifier 000457369
Title Knowledge representation for affective and adaptive tutoring systems
Alternative Title Αναπαράσταση γνώσης σε συναισθηματικά και προσαρμοστικά συστήματα μάθησης
Author Δούγαλης, Αχιλλεύς-Νικόλαος Β.
Thesis advisor Πλεξουσάκης, Δημήτριος
Reviewer Αναστασιάδης, Παναγιώτης
Αργυρός, Αντώνιος
Καλογιαννάκης, Μιχαήλ
Πάτκος, Θεόδωρος
Στεφανίδης, Κωνσταντίνος
Τσαμαρδινός, Ιωάννης
Abstract In recent years, there has been a meteoric rise in the use of educational technology. Intelligent Tutoring Systems (ITS) and Affective Tutoring Systems (ATS) are computer systems that can successfully teach courses to users at all levels of education, using AI techniques (the former) and Affect Sensing technologies (the latter) in order to facilitate learning. These systems are typically built around a specific subject such as algebra or genetics,etc, making reusability in different domains difficult. In contrast to ITS, Adaptive Learning Systems are more domain-agnostic, being able to teach more than one course. Adaptive Learning Systems concentrate on the presentation of a course by adapting it to the user’s learning preferences. These adaptations are based on learning theories that classify learners according to the way they learn; these theories are known as Learning Styles. Two major disadvantages of these systems is that they do not offer any type of feedback to the user (emotional or cognitive), and that there are doubts in the scientific community concerning the validity of Learning Styles Theories due to lack of relevant scientific evidence. Massive Online Open Courses (MOOCs) are the latest word in educational technologies; they deal with the problem of integrating many courses, users and instructors in a single platform. In order to be able to build better MOOC platforms, a course-teaching system should combine the advantages of ITS, ATS and Adaptive Learning Systems. Also, there is a need for a unified model that will be able to adequately represent a range of different courses. Finally this system should have the ability to calibrate its teaching strategies during interaction with users. This doctoral dissertation describes the design, implementation and evaluation of AffLog (affective Logic) Tutor, an ATS with the following specifications: The domain, tutoring and student models are designed using the Predicate Calculus and the Event Calculus. They are then implemented following the Answer Set Programming (ASP) formalism using the programming language Clingo. Also, the student model contains information about the user’s learning style according to the Felder-Silverman model. AffLog uses AI methods such as Planning and Projection in order to select the most suitable parts of the course for the current user according to the user's learning style and suitably present these parts to the user. The affective model of the system is designed to react to the user’s current emotional state providing advice and encouragement, thus facilitating the learning process. In order to evaluate the system, a simple course containing instructions on how to play the “Settlers of Catan” board game was designed and implemented. Evaluation of the system showed that users had high learning gains from their interaction with the system. Also, in an effort to find if Learning Styles are relevant to AffLog’s learning gains, a second ATS system, AffLogRL was developed. AffLogRL is similar to AffLog except that the latter replaces all modules that make use of Learning Styles with a Reinforcement learning agent. This agent uses the experience it accumulates by interacting with the users in order to formulate a teaching policy. After the training of AffLogRL with over 100 human users, the system was evaluated. The results showed that AffLogRL performed similarly to the AffLog system indicating that with more training and using Relational approaches the RL will perform better.
Language English
Subject Adaptive Learning Systems
Affective Tutoring Systems
Answer Set Programming
Learning styles
Reinforcement Learning
Ενισχυτική Μάθηση
Μέθοδοι μάθησης
Συναισθηματικά συστήματα διδασκαλίας
Συστήματα προσαρμοζόμενης εκμάθησης
Issue date 2023-07-21
Collection   School/Department--School of Sciences and Engineering--Department of Computer Science--Doctoral theses
  Type of Work--Doctoral theses
Permanent Link https://elocus.lib.uoc.gr//dlib/8/a/0/metadata-dlib-1689673272-696067-8458.tkl Bookmark and Share
Views 781

Digital Documents
No preview available

Download document
View document
Views : 2