Support for Collaborative Learning From Well-Thought-Out Group - PowerPoint PPT Presentation

Advancements in Intelligent Support for Collaborative Learning From Well-Thought-Out Group Formation to Effective Peer Interactions Seiji Isotani Associate Professor Department of Computer Systems University of Sao Paulo sisotani@icmc.usp.br

Team Affiliated PI Researchers Seiji Carla Leonardo Rafaela PhD Students Simone Cida Fernando Geiser Helena Rachel Master Students Bruno Armando Wilmax Kamila 2

3 Context The field of Computer-Supported Collaborative Learning - CSCL dedicates to study about how technology can be used to support collaborative learning and its processes (Stahl et al., 2006)

4 Context The field of Computer-Supported Collaborative Learning - CSCL dedicates to study about how technology can be used to support collaborative learning and its processes (Stahl et al., 2006) Despite of the potential benefits of Collaborative Learning, this approach is only beneficial when there is an adequate design and orchestration of its scenarios (Hernández-Leo et al., 2006, 2011; Dillenbourg, 2013)

5 Context Groups CL Design Group Formation The Problem Sequence of activities • These activities are too complex and time consuming ... • They also require specific knowledge and skills Teacher Learners Interaction Meaningful Support and Results Analysis

How to increase the collaborative learning (CL) ? chances of successful

How to provide intelligent carry out collaboration ? support to design and

Challenges !

Knowledge to design effective collaboration is distributed across several learning theories and pedagogical practices Isotani, S; Mizoguchi, et al. (2009) An ontology engineering approach to the realization of theory-driven group formation. International Journal of Computer-Supported Collaborative Learning , v. 4, p. 445-478.

They do not share the same terminology, assumptions and expectations and can be even contradictory! Mor, Y., Craft, B., & Hernández-Leo, D. (2013). The art and science of learning design. Research in Learning Technology , 21 . http://dx.doi.org/10.3402/rlt.v21i0.22513

Furthermore, if we consider only 15 pedagogical practices (3 dosage levels) , there are 205 trillion options to be considered Koedinger, K. R., Booth, J. L., & Klahr, D. (2013). Instructional complexity and the science to constrain it. Science , 342(6161), 935-937..

12 So, the question is ... Can we organize this pedagogial knowledge and build a computational infrastructure to use it adequately?

Our Approach Use ontological engineering W(A)-goal W(A)-goal W(A)-goal W(A)-goal How does the group How does the group Common goal Common goal Common goal change its state? change its state? W(L)-goal W(L)-goal to describe formally meaningful W(L)-goal W(L)-goal W(L)-goal G G Goal state Goal state Primary focus (P) Primary focus (P) Primary focus (P) k./cog. state (Group) k./cog. state (Group) What activity does the What activity does the Role Role Role Role Role S S information contained in theories group want to do? group want to do? Secondary focus (S) Secondary focus (S) Secondary focus (S) How does the learner How does the learner Role Role Role Role Role change his/her state? change his/her state? S<=P-goal S<=P-goal S<=P-goal Y  I-goal Y  I-goal Y  I-goal Y  I-goal Y  I-goal Y  I-goal I-goal I-goal I-goal I-goal G G P<=S-goal P<=S-goal P<=S-goal Goal state Goal state Y  I-goal Y  I-goal Y  I-goal Y  I-goal Y  I-goal Y  I-goal k./cog. state k./cog. state k./cog. state G G S S Y<=I-goal Y<=I-goal Y<=I-goal Y<=I-goal Y<=I-goal Y<=I-goal Y<=I-goal Y<=I-goal I-role I-role I-role I-role I-role I-role Behavior Behavior Behavior Behavior Behavior Behavior Why does the learner want to Why does the learner want to You-role You-role You-role You-role You-role You-role interact with other learners? interact with other learners? Behavior Behavior Behavior Behavior Behavior Behavior I-goal (I) I-goal (I) I-goal (I) I-goal (I) I-goal (I) I-goal (I) I-goal I-goal I-goal I-goal I-goal I-goal G G I-goal I-goal I-goal I-goal I-goal I-goal G G Pedagogical knowledge Ontological structure Run experimental studies to:  propose group formation;  design group activities; Use ontologies to  estimate benefits, etc.. support the development of ontology-aware systems users Teachers and students Theory aware intelligent systems 13

Formalizing CL interaction smaller group Whole group part of the whole L A L B L C 14

Formalizing CL Sub-group goal Strategy B Strategy A Whole group goal Role Role L A L B Role Individual goal Individual goal L C Individual goal 15

Formalizing CL Knowledge Formalization  Learning Strategies  Learning Goals  Group Goals Y<=I -goal(L A <=L B )  Roles Y<=I-goal (L B <=L A ) W(L)-goal({L A ,L B }) Role I-goal(L A ) Role I-goal(L B ) L A L B Role L C I-goal(L C ) W(L)-goal({L A ,L B ,L C }) 16

Interaction Patterns  Group Goals  Learning Strategies  Learning Goals Y<=I -goal(L A <=L B )  Roles Y<=I-goal (L B <=L A ) W(L)-goal({L A ,L B }) Role I-goal(L A ) Role I-goal(L B ) L A L B HOW? Role Interaction L C I-goal(L C ) W(L)-goal({L A ,L B ,L C }) Patterns 17

Interaction Patterns Interaction Patterns for Learning Theories proposed by Inaba et al. 2003 Ontological framework Cognitive Distributed Cognition Apprenticeship Influential I_L event Cognitive Peer Tutoring Constructivism I event Anchored Instructional event Sociocultural Theory Instruction Instructor Observational LPP Cognitive Flexibility L B Learning Theory Instructional action Action Interactions Benefits for the Instructor I-goal Learning Event Instructional Event L event Instructor Learner Learning event action LA Learner Role Role L A Learning object action LB L B L A Object Learning action learning goal (L A ) learning goal (L B ) Action Benefits for the Learner Influential I_L Events I-goal 18

CL Ontology CL Role Necessary Condition (b) Knowledge/cognitive state Desired Condition CL Scenario Knowledge/cognitive state Learning Strategy How to collaborate Y<=I-goal Behavioral Role I-role Influential I_L event Leaner Role Holder I event (a) You-role Instructional event Role Holder Learner Instructor Role Holder I-goal (I) (e) 1..* I-goal Instructional action Action CL process W(A)-goal Benefits for the Instructor I-goal Common goal (c) W(L)-goal L event Learning event How to interact Interaction Pattern Learner Role Holder Necessary Interaction Activity Learning object Influential I_L event (d) Object Complementary Interaction Activity Learning action Influential I_L event Action Benefits for the Learner 19 I-goal

CHOCOLATO: Concrete and Helpful Ontology-aware Collaborative Learning Authoring Tool Effective Groups CL Design Group Formation Ontologies Theories W(A)-goal W(A)-goal W(A)-goal W(A)-goal Common goal Common goal Common goal How does the group How does the group change its state? change its state? W(L)-goal W(L)-goal W(L)-goal W(L)-goal W(L)-goal G G Goal state Goal state Primary focus (P) Primary focus (P) Primary focus (P) What activity does the What activity does the Role Role Role Role Role S S k./cog. state (Group) k./cog. state (Group) group want to do? group want to do? Secondary focus (S) Secondary focus (S) Secondary focus (S) How does the learner How does the learner Role Role Role Role Role change his/her state? change his/her state? S<=P-goal S<=P-goal S<=P-goal Y  I-goal Y  I-goal Y  I-goal Y  I-goal Y  I-goal Y  I-goal I-goal I-goal I-goal I-goal Sequence of activities P<=S-goal P<=S-goal P<=S-goal G G Goal state Goal state Y  I-goal Y  I-goal Y  I-goal Y  I-goal Y  I-goal Y  I-goal G G S S k./cog. state k./cog. state k./cog. state Y<=I-goal Y<=I-goal Y<=I-goal Y<=I-goal Y<=I-goal Y<=I-goal Y<=I-goal Y<=I-goal I-role I-role I-role I-role I-role I-role Behavior Behavior Behavior Behavior Behavior Behavior You-role You-role You-role Why does the learner want to Why does the learner want to You-role You-role You-role interact with other learners? interact with other learners? Behavior Behavior Behavior Behavior Behavior Behavior I-goal (I) I-goal (I) I-goal (I) I-goal (I) I-goal (I) I-goal (I) I-goal I-goal I-goal I-goal I-goal I-goal I-goal I-goal I-goal I-goal I-goal I-goal G G G G ... Learners CHOCOLATO Interaction Meaningful Analysis results Isotani, S; Mizoguchi, et al. (2009) An ontology engineering approach to the realization of theory-driven 20 group formation. International Journal of Computer-Supported Collaborative Learning , v. 4, p. 445-478.

21 How to group students? ) ) Student 1 Student 2 ) Student 3

22 How to group students? ) ) Student 1 Student 2 ) Student 3

23 How to group students? ) ) Student 1 Student 2 ) Student 3

24 How to group students? ) ) Student 1 Student 2 ) Student 3