1. Hello world!

2. Hello world  I am Irene!
HCII since April, 2016!

3. Hello world  I am Irene! Greece!!!
PhD: Methods and tools for the evaluation of collaborative learning activities using time series (2014, University of Patras)

4. University of Pittsburgh (LRDC)
“Linking Dialogue with Student Modeling to Create an Enhanced Micro-adaptive Tutoring System”
University of Pittsburgh (LRDC)
&
CMU (HCII)

5. Dialogue as a means for learning
We aim to develop an adaptive tutorial dialogue system, guided by a student model that will support students in learning physics

6. Research questions Level of Support (LOS) Research Objective:
RQ1: What makes tutorial dialogues successful? Teachers’ adapt the level of discussion to the student’s “zone of proximal development” (Vygotsky)
RQ2: How do tutorial dialogues adapt to different student characteristics and prior knowledge?
Degree of Teacher Control (van de Pol)
Contingent Tutoring (Pino-Pasternak)
Cognitive Complexity (Nystrand, Graesser)
Research Objective:
What makes some tutor’s help generous or stingy, easy or challenging, straightforward or “cognitively complex”?
Level of Support (LOS)

7. An example would be nice….
RQ: What minimum acceleration must the climber …….
S: 500/55 kg=a m/s^2
T: I don't agree - that's the acceleration that just the pull from the rope would produce (well once the units are straightened out it would be). Think a little more. What is the general rule for finding acceleration from forces?
S: F/m=a
T: and what is the F there?
S: tension?
T: No.. the F in F=ma is always the net force on the object (or group of objects). The vector sum of all the forces on the object. I prefer to say "Sum of F= ma" because it's easier to get it right. So.. if she is sliding down and the rope is just short of breaking, what is the *net* force on her?
RQ: What minimum acceleration must the climber have in order for the rope not to break while she is rappelling down the cliff? (You do not have to come up with a numerical answer. Just solve for "a" without any substitution of numbers.)
S: a = f / m
T: what's f ?
S: f = mg
T: just mg ? how many forces act on the climber ?
S: mg + T
T: is mg down or up ?
S: down and T is up
T: ok so now solve for a again plugging in T and mg
High performer

8. Work in progress!
[How] can we group the features of instructional dialogue to differentiate the “levels of support” (LOS)?
Analyze human-to-human tutorial dialogues
Build a coding scheme to operationalize Level of Support
Set up Tutorial Dialogue Authoring Guidelines
Adapt tutorial dialogue system to LOS support
And then… what a great time for studies!

9. If you want to know more, get in touch!
The end.
If you want to know more, get in touch!
NSH 2602K
(plus cat pictures!)

10. Future plans
“The world is not enough” (J. Bond)

11. Method of the study 4 Coders LOS Coding Scheme
Level of Control [3-step scale]
Question Categories
[18 types]
Level of Specificity
[3-step scale]
Contingent Tutoring
[binary]
4 Coders
LOS
Coding Scheme
3 human-to-human dialogues on Physics / 1 per overall learning gains level

12. “Level of Support” LOS Coding Scheme
Dimension
Description
Coding
D1. Information provision (B/F)
The amount and level of specificity of the information ADDED by the tutor
3-step scale
(low/med/high)
D2. Hints provision
(F)
The hints/suggestions that are provided to the student. It should reflect the vagueness of the hints
D3. Feedback on correctness
(B)
the feedback the tutor provides to the correctness of the student’s previous answer
(None/Implicit/
Explicit)
D4. Information related to feedback
Additional information that is provided as explanation along with the tutor's feedback on correctness
Binary (Yes/No)

13. Dimension 1. Information related to the answer
Low
Medium
High
None / NA
Example: S: F/m=a” T: and what is the F there?
Example: S: F/m=a” T: and what is the F there? I prefer to say “Sum of F= ma” because it’s easier to get it right”
Example: S: F/m=a” T: and what is the F there? Fw is still 539 N down. So we have 539 N down and 500 N up. So what is the net force?”
 Example: S: F/m=a” T: ok
in the examples given above, the tutor expects from the student to mention the forces that make up the net force

14. Application and Evaluation
10 human-to-human tutorial dialogues (Physics) – 3 High, 3 Low, 4 Medium
2 raters per dialogue
The raters were given a tutorial on the coding scheme and detailed instructions
Dimensions
Cohen's kappa
D1. Information Provision (B)
0.871
D1. Information Provision (F)
0.843
D2 –Hints Provision
D3-Feedback on correctness
0.826
D4-Information related to feedback
0.764

15. So you coded it. Now what?
Initial Goal: Appropriate, adaptive dialogic support
What is appropriate for whom?
Let the experts tell! Great time for a study 
author dialogic support for various student types
Provide alternatives to existing tutorial dialogues
Ask for teachers’ input: what do you prefer?
what would you say? why?

16. Work in progress: Teachers’ input
3 scenarios [1 scenario per student type]
3 teachers
3 alternatives responses [authored, system’s, teacher’s free input]
Dialogue Set A
Dialogue Set B
Dialogue Set C
Low-Performer
Scenario 1
Scenario 3
Scenario 2
Medium-Performer
High-Performer

17. Dialogue Authoring Rules
Information and Explicit feedback to the Low Performers
Hints and Implicit Feedback to the High Performers
Low Performers : No, this is not right. If there is nothing to cause velocity to change (for example, some force), then the velocity will remain the same. Please list all forces that are applied on the arrow while it is in flight.
High Performers: Is there something that could cause velocity to change? What can this be?
Question: During the arrow`s flight, how does its horizontal velocity change (increases, decreases, remains the same, etc.)? Remember that you can ignore air resistance."
Student: decreases

18. Future work Author dialogues to support students of various levels
Implement a student model that will guide the adaptive dialogue tutoring system
Additive Factors (AFM), Performance Factors (PMF) and Instructional Factors (IFM)
semantic analysis: NLP algorithms for keywords/concept extraction