Simulated Student: Building Cognitive Model by Demonstration
Noboru Matsuda
School of Computer Science
Carnegie Mellon University
2 PSLC Summer Intern Seminar :: Noboru Matsuda :: July 11, 2006PSLC Summer Intern Seminar :: Noboru Matsuda :: July 11, 2006
Sim
ula
ted
Stu
den
t Pro
ject
CTAT
• Works great for the Example-Tracing Tutors
– A cognitive model specific to a particular problem
• For Model-Tracing Tutors, the author must build a generalized cognitive model from scratch
– Writing production rules is challenging
– Also need a cognitive task analysis
– Performing a task is much easier…
3 PSLC Summer Intern Seminar :: Noboru Matsuda :: July 11, 2006PSLC Summer Intern Seminar :: Noboru Matsuda :: July 11, 2006
Sim
ula
ted
Stu
den
t Pro
ject
Simulated Students
• Machine learning agent that …
– Observe model solutions
– Learn problem-solving steps
– Outputs a set of production rules
• Model of human students learning cognitive skills
• Programming by Demonstration
Lau, T. A., & Weld, D. S. (1998). Programming by demonstration
Blessing, S. B. (1997). A programming by demonstration authoring tool for model-tracing tutors
}Both for correct and incorrect steps
4 PSLC Summer Intern Seminar :: Noboru Matsuda :: July 11, 2006PSLC Summer Intern Seminar :: Noboru Matsuda :: July 11, 2006
Sim
ula
ted
Stu
den
t Pro
ject
Next Generation AuthoringBuild a tutor GUI
Demonstrate a solution
SimSt. learning
Rule simplify-LHS:
IF is-equation( Eq ),
is-lhs( Eq, Lhs ),
polynomial( Lhs ),
all-var-terms( Lhs )
Then simplify( Lhs, S-lhs ),
enter( S-lhs )
Production Rules
Rule simplify-LHS:
IF is-equation( Eq ),
is-lhs( Eq, Lhs ),
polynomial( Lhs ),
all-var-terms( Lhs )
Then simplify( Lhs, S-lhs ),
enter( S-lhs )
Rule simplify-LHS:
IF is-equation( Eq ),
is-lhs( Eq, Lhs ),
polynomial( Lhs ),
all-var-terms( Lhs )
Then simplify( Lhs, S-lhs ),
enter( S-lhs )
5 PSLC Summer Intern Seminar :: Noboru Matsuda :: July 11, 2006PSLC Summer Intern Seminar :: Noboru Matsuda :: July 11, 2006
Sim
ula
ted
Stu
den
t Pro
ject
6 PSLC Summer Intern Seminar :: Noboru Matsuda :: July 11, 2006PSLC Summer Intern Seminar :: Noboru Matsuda :: July 11, 2006
Sim
ula
ted
Stu
den
t Pro
ject
No Free Lunch
• Background knowledge– Operators
– Feature predicates
• External Jess function written in Java
(defrule multi-lhs…?var22140 <- (column (cells ? ? ?var22143 ? ? ? ? ?))?var22143 <- (cell (value ?val0&~nil))
(test (fraction-term ?val0 )) =>
(bind ?val2 (denominator ?val0))(bind ?input (mul-term-by ?val0 ?val2))…
)
7 PSLC Summer Intern Seminar :: Noboru Matsuda :: July 11, 2006PSLC Summer Intern Seminar :: Noboru Matsuda :: July 11, 2006
Sim
ula
ted
Stu
den
t Pro
ject
Example: Stoichiometry Tutor
Approximate Time Distribution (hours)
55
30
15
40
Operator generation
Feature predicate generation
Debugging
Testing
8 PSLC Summer Intern Seminar :: Noboru Matsuda :: July 11, 2006PSLC Summer Intern Seminar :: Noboru Matsuda :: July 11, 2006
Sim
ula
ted
Stu
den
t Pro
ject
Learn more about Simulated Students
http://www.cs.cmu.edu/~mazda/CTAT/SimSt/
Example Tutor (Equation)
User’s Guide
… and more would come soon
Send email to [email protected] (Noboru)