Computer Science 872 Detailed Information

Instructor

Gord McCalla (GM), Thorvaldson 281.4, 966-4902, <mccalla@cs.usask.ca>

Office hours: by appointment (to arrange a time send me an e-mail or chat to me after class)

Time and Place: Tuesday, 2:30 – 5:30, Spinks S372 (breakout room)

Jim Greer (JG), Julita Vassileva (JV), Ben Daniel (BD) will each take one week of the course; also, members of the ARIES laboratory will provide demonstrations of locally developed learning technologies

Text

Beverly Park Woolf, Building Intelligent Interactive Tutors, Morgan Kaufmann, 2009, 467 p.

Workload

Individual: some work will be done individually

20%: paper presentations (select a paper, summarize it, and present it in class) – there will be at least 2 such presentations, maybe more, depending on class size and other constraints

15%: participation (in class and on-line)

Group: there will be a group project, with marks as follows

10%: project proposal

5%: project proposal presentation

40%: final project

10%: final project presentation (including demonstration)

(note: the entire group will share equally in the group grades; but if there has not been equal sharing of effort among members of the group, differential marks may be awarded)

Description

“Aspects of advanced learning technology are studied, including: learner modelling, instructional planning, domain knowledge representation, authoring tools, tutorial dialogue, evaluation, semantic web technology, and theories of learning. The course takes an applied perspective, with the goal of understanding current resarch issues involved in building intelligent systems for use by learners.”

The course will be a bit more focussed than implied by this description and a bit more adventurous, too. The main goal will be to come to an understanding of the research area known as artificial intelligence in education (AIED). AIED is the area of e-learning that tries to combine research on the frontier of computer science (the “AI”) with research on the frontier of social science (the “ED”). AIED as a research area is extremely applied, but draws on both theoretical and applied ideas and techniques across a wide range of computer science and social science. It has been said that AIED is “AI-complete” in the sense that any problem in AI is instantiated within AIED, often in a way that makes it more interesting and tractable than the general AI problem. Thus, AIED can often explore research issues that in their general form are “too tough” for AI itself to deal with at present.

Outline

Date	Topic	Lect.	Comments
Sept. 7	Overview of the field: motivation, goals, history, background, the basic AIED architecture, example systems	GM
Sept. 14	Local ITS projects: iHelp, Recollect, OATS, ProTutor, Conundrum, Scottlebot	ARIES	GM away all week
Sept. 21	Domain knowledge representation, learner modelling, personalization	GM
Sept. 28	Papers (focus on knowledge), eg. cognitive tutor, Andes, SCENT, …	students
Oct. 5	Pedagogy, planning, interaction, communication	GM
Oct. 12	Papers (focus on interaction), eg. AutoTutor, TLTS, LISTEN, …	students
Oct. 19	Alternative pedagogical styles: collaborative environments, goal-based scenarios, situated cognition, virtual learning environments, distributed communities of practice	GM
Oct. 26	Papers (focus on alternative pedagogies), eg. Betty’s Brain reciprocal learning system, Lester’s interactive narratives, Fischer’s culturally and socially sensitive systems, Schank’s goal-based scenarios, Soloway’s science discovery environments, …	students
Nov. 2	Project proposal presentation: special time, later in the week	students	GM away early week
Nov. 9	Evaluation (how to evaluate an advanced learning system): experimental design, statistical techniques, data mining	BD, GM
Nov. 16	Julita Vassileva’s advanced learning technology	JV
Nov. 23	Jim Greer’s advanced learning technology	JG
Nov. 30	Gord McCalla’s advanced learning technology: issues papers: fragmented learning, active learner modelling, ecological approach. Following up on these issues: learning companions (Chan, Aimeur), lifelong learning (Kay, Fischer), educational data mining (Baker, Beck)	GM
Dec. 7	Project presentation and demonstration: special time	students

Note: this outline is subject to change as the course proceeds