GRAPPLE: Generic Responsive Adaptive Personalized Learning Environment

1
GRAPPLEGeneric Responsive Adaptive Personalized
Learning Environment

• Presented by Alexandra Cristea, University of
  Warwick

2
Outline of GRAPPLE Presentation

• Facts, TEL Challenges addressed, expected impact,
  projects
• Aims, related projects, development goals,
  partners and complementarity, vision, website
• Requirements analysis and demos
• IMS-LD and Adaptation
• Authoring
• User modeling
• Management
• QA

3
GRAPPLE facts

• Generic Responsive Adaptive Personalized Learning
  Environment
• 3 Year STREP
• Start date 1 February 2008
• Grant 3.850.000
• Number of partners initial 14, now 15 9
  countries
• Project coordinator Technische Universiteit
  Eindhoven, Netherlands
• Website http//www.grapple-project.org/
• FP challenge and objective EU FP7 STREP Project
  ICT Work Programme TEL Technology enhanced
  learning research is part of Challenge 4,
  'Digital Libraries and Content'.
• ftp//ftp.cordis.europa.eu/pub/ist/docs/telearn/te
  learn-fp7-projects_en.pdf

4
Challenges Research objectives

• The work programme aims at stimulating research
  that will lead to the creation of responsive
  environments for technology-enhanced learning
  and, at longer-term, the development of adaptive
  and intuitive learning systems.
• Responsive environments for technology-enhanced
  learning should be pedagogically sound and
  compatible with business processes and human
  resources management systems of organisations.
  They should exploit, where relevant,
  interactivity and collaboration of learners. In
  particular, they should be able to
• motivate, engage and inspire learners,
• accommodate personalisation to respond to
  specific learning needs and contexts (mass
  individualisation)
• support the transformation of learning
  outcomes into permanent and valuable knowledge
  assets
• enhance competence, skills and performance.
• Adaptive and intuitive learning systems,
  resulting from longer-term research efforts, will
  be able to configure themselves according to
  their understanding and experience of learners'
  behaviour. These systems will
• identify learner's requirements from
  monitoring progress in an intelligent way,
• make best use of the individual learning and
  cognitive abilities of the learner,
• give meaningful advice to both learners and
  teachers.

5
WP Expected impact

• faster and more effective learning, acquisition
  of knowledge, competences and skills
• unlocking people's and organisations' ability to
  master knowledge and apply it
• increased knowledge worker productivity
• more efficient organisational learning processes.

6
TEL FP7 projects
ftp//ftp.cordis.europa.eu/pub/ist/docs/telearn/te
learn-fp7-projects_en.pdf

• 80Days - Around an Inspiring Virtual Learning
  World in Eighty Days
• GRAPPLE - Generic Responsive Adaptive
  Personalized Learning Environment
• idSPACE - Tooling of and Training for
  Collaborative, Distributed Product Innovation
• LTfLL - Language Technologies for Lifelong
  Learning
• MATURE - Continuous Social Learning in Knowledge
  Networks
• SCY - Science Created by You

7
GRAPPLE - Generic Responsive Adaptive
PersonalizedLearning Environment

• aims at delivering a technology-enhanced learning
  environment for
• life-long learning, able to automatically adapt
  to personal preferences, prior knowledge, skills
  and competences, learning goals and the personal
  or social context in which the learning takes
  place.

8
GRAPPLE (a few) other (FP6) projects

• APOSDLE Scruffy Approach learn_at_work
• areas used by G. context-based retrieval user
  profiles maintenance
• LUISA SW Services Approach LO interchange by
  knowledge workers
• For G. link to competencies and skill-based
  adaptation
• PROLIX Proc.-oriented learning life cycle
  Approach
• G. will add adaptation services
• WINDS Web based intelligent design tutoring
  system
• G. builds on personalized recommender services
• RAFT Remote Accessible Field Trips
• G. can use personal annotations, metadata, and
  authentic and contextualized learning materials
• PROLEARN NoE in Professional Learning
• G. use requirements, adaptive methods, standards
  for adaptation specification and exchange of
  adaptive learning materials interoperability and
  authoring
• TENCompetence NoE Lifelong Competence
  Development
• G. use service based infrastructure
  integration of informal/formal learning support
• MACE Metadata for Architectural Contents in
  Europe
• G use personalized and contextualized
  perspectives on learning materials
• ALS Adaptive Learning Spaces
• G. use AHA!, initial Sakai connection, MOT new,
  standards compatibility (IMS-CP, IMS-QTI), new
  LAG, MOT2AHA

9
Developing

• A common abstract description of desired adaptive
  behaviour, using conceptual adaptation models,
  and a 'translation' to lower level adaptation
  rules that can be used by actual adaptation
  engines.
• Authoring tools for CAM, capturing relationships
  between concepts that are of a navigational
  and/or pedagogical nature, e. g. prerequisite
  relationships.
• extensions to standards
• A general purpose adaptation engine that can
  adapt any (xml) information
• User modelling services a distributed user
  modelling architecture
• Integration in different existing learning
  management systems, open source (Moodle,
  Claroline, Sakai) or proprietary (realised by
  GRAPPLE's industry partners).
• evaluation framework usability and effectiveness
  of adding adaptive behaviour to learning
  applications, in higher education and in
  corporate settings.
• Documentation and training material, mostly for
  authors of adaptive learning material
  http//www.learning-demo.eu/

10
user modeling architectures
adaptive e-learning
TU/e
TUD
DFKI
LUH/L3S
authoringmetadata
interfaces /server technology
VUB
Warwick
OUNL
TCD
standardslearning design
interaction
ATOS
USI
GILABS
UCL
UCAM
IMC
UniGraz
industrial learning technology
open source learning management
evaluation
11
The GRAPPLE Vision/Objectives

• define and realize a generic adaptation engine,
  unifying current research ideas (and prototypes)
• define and realize a generic user modeling
  component (based on current research ideas)
• define and build a broker architecture between
  multiple adaptive applications and user model
  services
• integrate this infrastructure into existing open
  source and proprietary academic and industrial
  learning management systems.

12
(No Transcript)
13
D9.1 D10.1 Requirements analysis

• Goal elicitation of stakeholders (learners,
  training providers) expectations and needs to
  guide GRAPPLE development
• Method
• Application scenarios
• Conduction of interviews
• Consolidation and analysis of interviews

14
Conduction of interviews

• Interviewing of
• 8 corporate / 8 academic learners
• 18 corporate / 17 academic training providers
• 2 others
• Creation of a summary and data sheet for each
  interview
• Analysis of the interview data in terms of
  requirements analysis to elicit users ideas,
  expectations and needs on ALEs

15
Demo course educational game
http//learning-demo.eu/aha/BE/
16
What is an essential functionality of an
adaptive learning environment ?

• Consideration of previous knowledge and
  competencies
• Content personalization
• Authoring advice (suggestions for teachers and
  authors)
• Evaluation and monitoring features
• Ease of use
• Transparent adaptation process

17
Remarks from the Interview Summaries

• Variation in experience wrt adaptive learning
  systems
• Appreciation of the complexity of the authoring
  task
• Limited resources
• Profit gt Effort
• Transparent adaptation
• Ability to scrutinize
• Learn to learn

18
Assessment of adaptation technologies by
interviewees for industry
19
GRAPPLE ALE LMS components - integration

• Domain Model (WP3)
• Conceptual Adaptation Model (WP3)
• User Modelling Storage (WP6)
• User Model (WP2)
• Single sign-on facility Shibboleth
• LMS
• Claroline
• Moodle
• Sakai
• Commercial partners LMSs

28 July 2008
GRAPPLE meeting, Hannover
Slide 19
20
IMS LD and Adaptation

• IMS LD can represent some adaptation methods, but
  not all of them
• Specification of concrete learning design
  instances is usually context dependent and does
  not support reusability very well
• Representation of various types of knowledge and
  their interaction to assist authors and to
  generate concrete instances dynamically
• Interoperability demands between systems
  between different models/layers
• Learning standards are not harmonized Semantic
  Web is used as mediator
• Authoring of learning design and adaptation
  strategies

21
WP3 Authoring
22
MOT Domain attribute creation
Domain concept attribute
Current domain concept
23
MOT Goal model creation
Ordering of lessons
Weights of sublesson
Labels of sublesson
24
LAG grammar

• PROG ? DESCRIPTION VARIABLES INITIALIZATION
  IMPLEMENTATION
• DESCRIPTION ? // text
• VARIABLES ? // text
• INITIALIZATION ?initialization ( STATEMENT )
• IMPLEMENTATION ? implementation ( STATEMENT )
• STATEMENT ? IFSTAT WHILESTAT FORSTAT
  BREAKSTAT GENSTAT SPECSTAT (STATEMENT)
  STATEMENT ACTION
• IFSTAT ? if CONDITION then (STATEMENT) if
  CONDITION then (STATEMENT) else (STATEMENT)
• WHILESTAT ? while CONDITION (STATEMENT)
  TARGETLABEL
• ACTION ? ATTRIBUTE OP VALUE
• CONDITION ? enough((CONDITION), VALUE) PREREQ
• PREREQ ? ATTRIBUTE COMPARE VALUE
• ATTRIBUTE ? GENCONCEPTATTR SPECCONCEPTATTR
• SPECCONCEPTATTR ? \SPECCONMAP\SPECCON\SPECATTR\AT
  TR.ATTRATTR
• LAOSCONCEPTMAP ? DM GM UM PM CM
• ATTR ? Attribute title keywords text
  introduction conclusion
• exercise child parent
  Relatedness ATTR.ATTR CONCEPT.ATTR label
  weight text
• ATTRATTR ? type order next ToDo menu
  show access visited text

25
Overall structure of the LAG adaptation strategy

• // Description
• // Variables
• initialization (
• // what the user sees first
• )
• implementation (
• // how the user interacts with the system
• )

26
LAG snippet

• // show to intermediate only what is relevant
• if enough(
• GM.Concept.label beginner
• UM.GM.stereotype1 beginner
• ,
• 2 ) then
• (
• PM.GM.Concept.show true
• )
• // change from beginner to intermediate if it is
  appropriate
• if (UM.GM.beginnerno 0) then
  UM.GM.stereotype1 int

27
GRAPPLE Domain Model

• Concepts and relations (hierarchical, is_a,
  part_of, etc.)
• Using a content repository

28
GRAPPLE CRT definition (pedagogic relationship
types)
29
GRAPPLE Example Graphical Authoring
Using generic placeholders
30
GRAPPLE Example Graphical Authoring
Conditions
31
Whats in the framework
32
Push/Pull Management Structure

• PMT Project Management Team
• push TU/e, leader of WP1 developing the generic
  adaptation engine
• pull OUNL and IMC, leaders of WP9 and 10 doing
  the evaluation in higher-education and in
  industry
• PMT Technical Quality Assurance Task Force
• same people will develop a QA plan and monitor
  its execution it will also monitor the
  peer-reviewing process

33
Other Management Structures

• EPMB Executive Project Management Board
• one representative from each partner
• takes all important decisions (including
  financial resources, adding/removing/replacing
  partners)
• Workpackage Management Teams
• one per WP, one person per partner involved in
  the WP
• chaired by WP leader
• day to day management of the WP
• responsible for turning in deliverables on time
• responsible for Quality management/control within
  the WP

34
SURFgroepen
Offered by the Dutch joint support
organization for universities and colleges (Surf)

• Video Conferencing Tool (Adobe Connect)
• Supports
• Video and Sound
• Audio All at once or one at a time
• Desktop / Application sharing
• Presenting powerpoint files
• Recording and Playback of meetings
• Whiteboard
• Chat
• Etc.

35
(No Transcript)
36
Mailing Lists

• Mailing lists for all work packages, evaluation
  work packages combined, general, pmt and
  administrative
• History available

37
EU-xpert

• Periodic Activity Reports
• Summarize the work done for the tasks /
  deliverables
• Work package leader tasks
• Distributes works to main scientific personal for
  every partner
• Needed for reporting to European Commission

38
(No Transcript)
39
EU-Fin

• Periodic financial reports
• Declare actual costs
• And compare with estimated costs
• Financial people tasks
• Needed for every partner for every work package
• Needed for financial reporting to European
  Commission
• Instead of Form C
• Saves a lot of time compared with paper form C

40
(No Transcript)
41
SVN

• Subversion version control repository
• Code repository for all GRAPPLE related code

42

• Thank you for your attention !
• Questions?