Please enable JavaScript.
Coggle requires JavaScript to display documents.
Concept of an Intelligent Tutoring System for the E-Learning Platform…
Concept of an Intelligent Tutoring System for the E-Learning Platform ArchiLab
Research
Problem Definition
Derived research question
How can adaptive learning systems help students recognize and improve bottlenecks in their learning?
How can intelligent tutoring systems improve learning strategies for students in an e-learning context?
What learning strategies for students can teachers design to improve the effectiveness and efficiency in an e-learning context?
What challenges in an e-learning context can intelligent tutoring systems help overcome?
What barriers in the knowledge transfer between teachers and students can intelligent tutoring systems help overcome?
How can intelligent tutoring systems help design high quality learning strategies and content for students in an e-learning context?
How effective can intelligent tutoring systems help e-learners bridge their knowledge and skills gaps?
one-on-one instruction?
What knowledge, skills, and expertise models can intelligent tutoring systems develope for designing high-quality learning strategies?
How can intelligent tutoring systems further individualize learning models for students in an e-Learning context?
How can intelligent tutoring systems help novices achieve expert competence level?
References
Papers
“
Intelligent Tutoring Systems: Using AI to Improve Training Performance and ROI
” by Ong and Ramachandran
knowledge models
expert model
student model
instructor model
Socratic teaching method
simulated procedures
competence gaps
ITS authoring tools
SimBionic
cost-effective one-on-one instruction
"
Intelligent Tutoring Systems
" by John R. Anderson
computer-assisted instruction
intelligent computer-assisted instruction (ICAI)
Major effort of ITS: "help students manage working memory load"
encode information on screen student is likely to forget
Active Control of Thought (ACT) Theory
four features
productions
immediate feedback about errors
model tracing
the
bug catalog
ideal model
goal structure
working memory limitations
knowledge compilation
instruction vs problem-solving?
"
Designing an Intelligent Tutoring System Based on a Reactive Model of Skill Acquisition
" by Randall W. Hill, Jr.
model tracing
ACT* (Anderson, 1983)
PUPS (Anderson, 1989; Anderson, 1990)
curriculum design choice
cognitive model = artificial student?
Soar
an architecture that implements a theory of human cognition
task skills
reactive
goal-oriented
problem space hierarchy
Soar problem spaces
#
knowledge compilation
What? Soar chunkin mechanism builds productions that summarize the subgoal results.
Techniques
ACT*
"new productions are built to summarize the computations performed while solving a problem"
PUPS
productions are built from the induced implications formed uring the analogy process or by composing two or more of these implications
cognitive modeling results for tutoring system design
Learning by doing
Learning about Preconditions
Rote Learning is Incomplete
optimal learning takes place in partial rote learning and performing the task and resolving impasses as they arise
Learning Occurs in a Goal Context
intervene tutorially in the context of an impasse
how to
recognize
and
explicate
problem solving impasses (Hill & Johnson, 1993)
Learning via Failure Impasse
how to capitalize on error recovery situations
Detecting Hidden Knowledge Gaps
Recovery from Incorrect Knowledge
"
Adaptive Systems: from intelligent tutoring to autonomous agents
" by David Benyon and Dianne Murray
alleviate variance in human-based teachings skills
targeted instruction in a constrained subject domain
user's knowledge state, identified expert's knowledge ("goal state")
current state of a learner's knowledge (Dede, 1986)
overall context of courses, syllabi and tutorial objectives (Self, 1987)
student model
knowledge about concepts and relationships
level? achievement? performance?
student performance held in user profile
expert model
tutor model
classified types of knowledge which determine the overall performance of a teaching program (Self, 1985)
teaching stategy
intelligent computer coach vs the Socratic tutor ?
discovery-learning vs guided discovery-learning and taks-centred learning-by-doing
Aim: "
ensure that a student attains a certain level of competence in a well-specified domain
"
Books
"
Advances in Intelligent Tutoring Systems
" by Roger Nkambou
Chapter 1: Introduction
overview of the field, different ITS architectures
(Nwana 1990)
Bloom (1984) proved tutor effectivity
Chapter 2: Modeling the Domain
Merrill’s Component Display Theory → Merrill (1991)
Bloom (1975); Gagne (1985)
clear classifications of knowledge and skills
expert module
the black box model
the glass box model
the cognitive model
mimics the human mind (Corbett et al., 1997)
Adaptive Control of Thought (ACT-R)
--> Anderson (1996)
e.g. Algebra Tutor, Geometry Tutor, LISP Tutor
production rules
#
production rules (Anderson, 1982)
rule based modeling
#
Forword
educational data mining community (EDM 2010)
User's goals and his or her broader context
Vassileva et al (2001); McCalla (2004)
"forcing function"
Artificial Intelligence in Education (AIED) --> research area
Intelligent Tutoring Systems (ITS) ..> kind of systems researchers build
association rule mining (Jannach et al 2011)
Topic
Designing Learning Models with Intelligent Tutoring Systems in an E-Learning Context
Providing Intelligent Computer-Assisted Instruction Concept for the E-Learning Platform ArchiLab
A Virtual Tutor Concept for High-Quality Immediate Learning Feedback in ArchiLab
Concept of an Intelligent Tutoring System for the E-Learning Platform ArchiLab
Document Structure:
Title Page
Declaration
Abstract
Table of Contents
List of figures
List of tables
Nomenclature
Chapters
Acknowledgements
References
Appendix
Chapter 1: Overview of the Field
different ITS architectures
traditional ITS architectures
Models
Chapter 2: