Faculty of Computer Science and Information Technology >
PhD Theses : Computer Science >
Please use this identifier to cite or link to this item:
|Title: ||Integrating Cognitive Agents for Problem-based Learning in an IT environment|
|Authors: ||Akcell Chii Chung Chiang|
|Keywords: ||Problem-based learning|
Information technology education
|Issue Date: ||2008 |
|Abstract: ||Problem-based Learning (PBL) has been widely used in medical or nursing education since the 1960s and is currently a major learning strategy for a variety of programs ranging from kindergarten to university. Organizations such as the World Health Organization (WHO) and World Bank recommend the use of PBL in education curricula.
IT education is one of the fastest changing industries; similar to medical or nursing education, IT courses have a high content load in teaching in order to respond to the fast-changing pace of IT industries. However, the large student numbers and limited staff in IT classes often hinder interested educators combining PBL discussion with traditional teaching methods in their curricula. Without an effective and efficient tool, it seems impractical to ask an educator to handle too many PBL groups in IT classrooms. We hypothesize that developing a cognitive software agent system can reduce the time pressure in teaching for educators and overcome the shortage of tutors or assistants in facilitating many PBL groups. Based on the hypothesis, our study seeks to answer the research question: “How can IT facilitate PBL discussion?” We propose an agent technology to implement the various components of PBL tasks. We investigate the difficulties of implementing PBL in IT courses and seek possible solutions to overcome the shortage of tutor. Along with the research, the study proposed a cognitive model with critical-thinking, and developed three agent algorithms to work as the tutor in PBL. The software agents are built on the notions of threshold and knowledge-weight from the discipline of machine learning. They encourage students to judge or criticize the solutions posted by others before exploring further knowledge-content. The cognitive system then sums up the judgment scores as its knowledge-weight in order to pass the thresholds set up for ranking/arranging the learning issues. We develop an agent system to work as a tutor to interact with learners without unnecessary educator’s involvement. There are three different agents integrated together for the PBL system: the GDSS-agent settles students’ conflicts in meetings; the tutor-agent consults students during the learning; the assessment-agent evaluates individual student’s formative and summative progresses in PBL classes.
In an IT classroom experiment, we successfully set up the PBL paradigm with MALESAbrain system facilitation. The system evaluations from both educators and students have confirmed that MALESAbrain can help them to improve the teaching and the learning. MALESAbrain experiment in our AI course has proved that the educators do not have any difficulty to handle many PBL groups, and the students can come out with good learning results. The experiment has shown the system and the adapted PBL pedagogy is able to meet the educators’ demands to complement the weakness of the traditional teaching methods for the IT classes.|
|Appears in Collections:||PhD Theses : Computer Science|