published_or_final_version / Computer Science and Information Systems / Master / Master of Philosophy

No description available.

No description available.

No description available.

No description available.

No description available.

Intelligent Tutoring Systems (ITSs) have been shown to significantly improve students' learning in a variety of domains, including physics, mathematics, and thermodynamics. Tutorial dialogues is one of the strategies used by ITSs and has been empirically shown to significantly improve learning. This project investigates how different students interact with the tutorial dialogues in EER-Tutor, using both eye-gaze data and student-system interaction logs. EER-Tutor is a constraint-based ITS that teaches conceptual database design. In order to have a more comprehensive and accurate picture of a user's interactions with a learning environment, we need to know which interface features s/he visually inspected, what strategies s/he used and what cognitive efforts s/he made to complete tasks. Such knowledge allows intelligent systems to be proactive, rather than reactive, to users' actions. Eye-movement tracking is therefore a potential source of real-time adaptation in a learning environment. Our findings indicate that advanced students are selective of the interface areas they visually focus on whereas novices waste time by paying attention to interface areas that are inappropriate for the task at hand. Novices are also unaware that they need help with understanding the domain concepts discussed in the tutorial dialogues. We were able to accurately classify students, for example as novice or advanced students, using only eye-gaze or EER-Tutor log data as well as a combination of EER-Tutor and eye-gaze features. The cost of eye-tracking is justified as classifiers using only eye-gaze features sometimes perform as well as those utilising both EER-Tutor and eye-gaze data and outperform classifiers using only EER-Tutor data. The ability to classify students will therefore allow an ITS to intervene when needed and better guide students' learning if it detects sub-optimal behaviour.

Empirical studies have shown that Intelligent Tutoring Systems (ITS) are effective tools for education. However, developing an ITS is a labour-intensive and time-consuming process. A major share of the development effort is devoted to acquiring the domain knowledge that accounts for the intelligence of the system. The goal of this research is to reduce the knowledge acquisition bottleneck and enable domain experts to build the domain model required for an ITS. In pursuit of this goal an authoring system capable of producing a domain model with the assistance of a domain expert was developed. Unlike previous authoring systems, this system (named CAS) has the ability to acquire knowledge for non-procedural as well as procedural tasks. CAS was developed to generate the knowledge required for constraint-based tutoring systems, reducing the effort as well as the amount of expertise in knowledge engineering and programming required. Constraint-based modelling is a student modelling technique that assists in somewhat easing the knowledge acquisition bottleneck due to the abstract representation. CAS expects the domain expert to provide an ontology of the domain, example problems and their solutions. It uses machine learning techniques to reason with the information provided by the domain expert for generating a domain model. A series of evaluation studies of this research produced promising results. The initial evaluation revealed that the task of composing an ontology of the domain assisted with the manual composition of a domain model. The second study showed that CAS was effective in generating constraints for the three vastly different domains of database modelling, data normalisation and fraction addition. The final study demonstrated that CAS was also effective in generating constraints when assisted by novice ITS authors, producing constraint sets that were over 90% complete.

Across many domains, Intelligent Tutoring Systems (ITSs) are used to facilitate practice, providing a customized learning environment and personal tutoring experience for students to learn at their own pace through effective student modeling and feedback. Most current ITSs are built around cognitive learning theories including Ohlsson's theory on learning from performance errors and Anderson's ACT theories of skill acquisition which focus primarily on providing negative feedback or corrective feedback, facilitating learning by correcting errors. Research into the behavior and methods used by expert tutors suggest that experienced tutors use positive feedback quite extensively and successfully. This research investigates positive feedback; learning by capturing and responding to correct behavior, supported by cognitive learning theories. The research aim is to develop and implement a systematic approach to delivering positive feedback in Intelligent Tutoring Systems, in particular SQL-Tutor, a constraint-based tutor which instructs users in the design of Structured Query Language (SQL) database queries. An evaluation study was conducted at the University of Canterbury involving a control group of students who used the original version of SQL-Tutor giving only negative feedback and an experimental group using the modified version of SQL-Tutor where both negative and positive feedback were given. Results of the study show that students learn quite similarly from one system to another, however those in the experimental group take significantly less time to solve the same number of problems, in fewer attempts compared to those in the control group. Students in the experimental group also learn approximately the same number of concepts as students in the control but in much less time. This indicates that positive feedback results in increased amount of learning over a shorter period of time and improves the effectiveness of learning in ITSs.

As one of the central problems in the area of Intelligent Tutoring Systems (ITSs), student modelling has been widely used to assist in systems’ decision making and students’ learning. On the one hand, by reasoning about students’ knowledge in the instructional domain, a system is able to adapt its pedagogical actions in order to provide a customized learning environment. These actions may include individualized problem-selection, tailored instructions and feedback, as well as updating the presentation of student models. On the other hand, students can reflect on their own learning progress by viewing individual Open Student Models (OSMs) and enhance their meta-cognitive skills by learning from the system’s estimation of their knowledge levels. It is believed that making the information in the student model available to students can raise students’ awareness of their strengths and weaknesses in the corresponding domain and hence allow them to develop a more effective and efficient way of learning. An OSM has been developed in EER-Tutor. EER-Tutor is a web-enhanced ITS that supports university students in learning conceptual database modelling. Students design Enhanced Entity-Relationship (EER) diagrams and receive different level of feedback in a problem-solving environment. The pedagogical decisions on feedback generation and problem selection are made according to student models. Previously, student models in EER-Tutor are presented to students on request as skill meters. Skill meters have been proved useful in helping students to improve their meta-cognitive skills. However, as the simplest presentation of a student model, skill meters contain very limited information. Some studies show that an OSM with multiple views is more effective since it supports individual preferences and different educational purposes. The focus of our work is to develop a variety of presentations for the OSM in EER-Tutor. For this purposes, we have modified the system to include not only skill meters but also other presentation styles. An evaluation study has been performed after the development. Both subjective and objective results have been collected. This thesis presents the extended EER-Tutor, followed by the analysis of the evaluation study.