Slide 1: An adaptative framework for tracking Web-based Learning Environments
Valentin Butoianu, Phillipe Vidal, Julien Broisin
Institut de Recherche en Informatique de Toulouse, France {butoianu, vidal, broisin}@irit.fr
Exploitation of Usage and Attention Metadata (EUAM 09), 28 September 2009, Lübeck, Germany
Slide 2: Context
Mass personalization
◦ Students: adapt learning scenarios to each user ◦ Instructors and designers: adapt their courseware to various users’ learning paths
=> Need to collect a large amount of attention metadata from Web-based Learning Environments
2
Exploitation of Usage and Attention Metadata (EUAM 09), 28 September 2009, Lübeck, Germany
Slide 3: Research issues
Existing tracking systems
◦ Students’ online activities within LMS • Uniform representation of ◦ Discussion forums attention metadata ◦ Chat, instant messaging, …
• drawbacks: Main An architecture promoting the share ◦ Specificityand reuse of attention ◦ Attention metadata enclosed into systems metadata
being traced
Prevents attention share, thus mass personalization
Exploitation of Usage and Attention Metadata (EUAM 09), 28 September 2009, Lübeck, Germany
3
Slide 4: Plan
1.Uniform representation 2. An opened and distributed architecture 3.Experimentation with Learning Objects 4.Conclusions and future works
Exploitation of Usage and Attention Metadata (EUAM 09), 28 September 2009, Lübeck, Germany
4
Slide 5: Our approach
Model based approach to represent attention metadata produced by various learning systems
◦ High level of abstraction ◦ Extensible
Reuse of WBEM, a « de facto » standard proposed by DMTF Dedicated to the network, systems and applications management.
Exploitation of Usage and Attention Metadata (EUAM 09), 28 September 2009, Lübeck, Germany
5
Slide 6: Modeling systems and resources
Exploitation of Usage and Attention Metadata (EUAM 09), 28 September 2009, Lübeck, Germany
6
Slide 7: Modeling users’ activities
Exploitation of Usage and Attention Metadata (EUAM 09), 28 September 2009, Lübeck, Germany
7
Slide 8: Plan
1.Uniform representation 2. An opened and distributed architecture 3.Experimentation with Learning Objects 4.Conclusions and future works
Exploitation of Usage and Attention Metadata (EUAM 09), 28 September 2009, Lübeck, Germany
8
Slide 9: Architecture
LEARNING ENVIRONMENT SOAP INT AGENT AGENT AGENT MIDDLEWARE TRACKING ENVIRONMENT
Learning Application 1
TRACKING SERVICE
WBEM INTERFACE
SOAP INTERFACE
SOAP HTTP
TRACKING MANAGER
SOAP INT
Learning Application 2
SOAP INT
Learning Application 3
SOAP HTTP
WBEM INTERFACE
SOAP INTERFACE
MODEL MANAGEMENT SERVICE
SOAP INT
Management Application
SOAP HTTP
TRACKING REPOSITORY
Exploitation of Usage and Attention Metadata (EUAM 09), 28 September 2009, Lübeck, Germany
9
Slide 10: Model management service
Dedicated to attention classes Easy specialization of generic models to represent any attention data
◦ Define new classes and/or properties ◦ Modify or delete existing classes and/or properties
Rules:
◦ Apply to classes inheriting only from the generic models ◦ Deny modifications of the generic models
Exploitation of Usage and Attention Metadata (EUAM 09), 28 September 2009, Lübeck, Germany
10
Slide 11: Tracking service
Dedicated to attention instances SetTrace method: self-adapt to the evolution of the model by creating a XML This methods are selfSchema adapting toto the model. conform the attention
model
GetTrace method: explores the tracking repository in order to get an unified view of all attention metadata
11
Exploitation of Usage and Attention Metadata (EUAM 09), 28 September 2009, Lübeck, Germany
Slide 12: Plan
1.Uniform representation 2. An opened and distributed architecture 3.Experimentation with Learning Objects 4.Conclusions and future works
Exploitation of Usage and Attention Metadata (EUAM 09), 28 September 2009, Lübeck, Germany
12
Slide 13: Modeling specific systems and resources
Exploitation of Usage and Attention Metadata (EUAM 09), 28 September 2009, Lübeck, Germany
13
Slide 14: Learning objects’ activities
Exploitation of Usage and Attention Metadata (EUAM 09), 28 September 2009, Lübeck, Germany
14
Slide 15: Specific architecture
LEARNING ENVIRONMENT SOAP INT AGENT MIDDLEWARE TRACKING ENVIRONMENT
Learning Application 1
SOAP HTTP
TRACKING SERVICE
WBEM INTERFACE
SOAP INTERFACE
OPEN PEGASUS
TRACKING MANAGER
AGENT
SOAP INT
Learning Application 2
gSOAP
WBEM INTERFACE SOAP INTERFACE MODEL MANAGEMENT SERVICE SOAP INT
SOAP HTTP
Management Application
SOAP HTTP
TRACKING REPOSITORY
Exploitation of Usage and Attention Metadata (EUAM 09), 28 September 2009, Lübeck, Germany
15
Slide 16: Use case: Consultation of a LO from Moodle
Moodle
4.Consult document
Tracking Tracking 2.Return model classes service manager
3.Build XML Schema
1.Request model classes
Tracking repository
5.Build XML trace 6.Send XML trace
7.Validate XML trace 8.Build CIM Instances 9.Send CIM Instances 10.Insert instances into repository
Exploitation of Usage and Attention Metadata (EUAM 09), 28 September 2009, Lübeck, Germany
16
Slide 17: Plan
1.Uniform representation 2. An opened and distributed architecture 3.Experimentation with Learning Objects 4.Conclusions and future works
Exploitation of Usage and Attention Metadata (EUAM 09), 28 September 2009, Lübeck, Germany
17
Slide 18: Conclusions and future works
A framework to manage attention metadata
◦ Integrated within ubiquitous operating systems ◦ Easy to satisfy multiple objectives: model management service ◦ Facilitates share and reuse :
Externalisation of attention metadata Self - adapting tracking service (Windows XP, Linux)
Automated update of agents according to the model Use Manager to Manager communication Exploit the collected traces: intelligent tutoring systems, recommendation systems
Exploitation of Usage and Attention Metadata (EUAM 09), 28 September 2009, Lübeck, Germany
18
Slide 19: Questions
Exploitation of Usage and Attention Metadata (EUAM 09), 28 September 2009, Lübeck, Germany
19
