issues with agreement on ontologies, changes to ontologies, mass annotation, etc...
WEB 2.0 HAS SIGNIFICANT IMPLICATIONS FOR EDUCATION; PARTICIPATION, COLLABORATION WEB 2.0 HAS GIVEN US IMPLICIT STRUCTURE AND LIGHTWEIGHT KNOWLEDGE MODELLING - BUT IT CANNOT BE USED BY MACHINES TO DO SEARCHING, MATCHING, LET ALONE ADDRESS INTEROPERABILITY AMONG DATA SOURCES
Semantic Technologies in HE Seminar - Learning Societies Lab
Perspectives on Semantic
Technologies for Education
Thanassis Tiropanis, Hugh Davis, Dave Millard, Mark Weal, Su White, Gary Wills
Asma Ounnas, Marcus Ramsden, Faith Lawrens, Heather S. Packer, Daniel A. Smith
• JISC-funded project working with CETIS
Survey of semantic tools and services
Current adoption of semantic technologies in the UK higher
Roadmap of semantic technology adoption in the next 5 years
The SemTech team: The JISC CETIS Semantic The SemTech workshop
Hugh Davis Technology Working Group: participants:
Sheila MacNeill (CETIS) Colin Allison (University of St.
Faith Lawrens Andrews)
David Millard Lorna Campbell (CETIS)
Chris Bailey (University of Bristol)
Phil Barker (CETIS)
Asma Ounnas Helen Beetham
Liliana Cabral (Knowledge Media
Heather S. Packer Institute, Open University)
Simon Buckingham-Shum (Open Patrick Carmichael (University of
Marcus Ramsden University, UK) Cambridge)
Daniel A. Smith David Davies (University of Warwick) Tom Franklin (Franklin
Thanassis Tiropanis Michael Gardner (University Essex) Consulting)
Mark Weal Tony Linde (University of Leicester) David Kay (Sero Consulting)
Wilbert Kraan (CETIS) George Magoulas (London
Su White Knowledge Lab, Birkbeck College)
Sue Manuel (University of
Gary Wills Loughborough)
Uma Patel (City University)
Learning Societies Lab Alex Poulovassilis (London
Lou McGill Knowledge Lab, Birkbeck College)
Graham Wilson (LT Scotland) John Scott (University of Essex)
(ECS-University of Robin Wylie (LT Scotland)
Southampton) David Kernohan (JISC)
semantic tech in edu scenario?
A top-down approach
semantic tech in a web 2.0 world
• Soft semantics
‣ Meaning in formats that humans can process
‣ Lightweight knowledge modelling in Web 2.0 applications
• Hard semantics
‣ Meaning in formats that machines can process
‣ Processing is independent of domain speciﬁc schemas
What is the value of Semantic
Technologies in HE?
Is it time to re-think the value of
semantic technologies in HE?
• Web 2.0 promise for content generation annotation
• Value of lightweight knowledge modelling and reasoning
• The HE institutional challenges
• The learning and teaching challenges
• Linked data movement and the Web of data
HE institutional perspective
• Visibility of degree programmes and research output of HE
• Curriculum design
• Recruitment and retention of students
• Efﬁciency of accreditation
• Collaboration across departments and institutions through workﬂows
• Integration of knowledge capital, cross-curricular initiatives
• Transparency of data held by educational institutions
learning and teaching perspective
• Course creation and delivery workﬂows
• Group formation for learning and teaching activities
• Critical thinking and argumentation support
• Personal and group knowledge space construction
• Assessment, certiﬁcation and addressing of plagiarism
surveyed semantic technologies
Repositories, VLEs Technologies for
and Authoring tools Linked Data and
Collaborative Authoring and
Unobtrusive user observation
Annotation of resource sections
Visualisation of arguments
Collaborative domain modelling
Real time meeting capture
Searching and Matching tools
Associations between experts
Mining RDF from existing repositories
Discovery, selection, negotiation and
composition of LOs
Use of Semantic Web Services
Repositories, VLEs, Annotation tools
Collaboratively authored, open
repository of structured topics
Topics mined from other repositories
Accessible via open APIs
Distributed network of semantically
aware shared annotation services in
the form of RDF stores
Support for user-facing applications
RKBExplorer D2R Server
What is the value of semantic
questions following the survey
• Where are the education speciﬁc users in these applications?
• What is the use of ontologies in surveyed applications?
• Where is the value of reasoning in surveyed applications?
• What is the value of semantic technologies in practice?
surveyed semantic tech use
• Actors: Tagging
Teacher Shared Bookmarking
University Administrator OWL
System Administrator FOAF
Program/Module Co-ordinator SKOS
surveyed learning activity use
• Collaborative activities • T&L activities involving the
Team Building individual
Computer Mediated Information Gathering
Discussion Information Handling
Computer Mediated Experimentation
Role Play Information Publishing
Content Creation Simulation
Content Annotation Experiments
• Most of the identiﬁed HE challenges can be addressed by querying
across institutional repositories (databases, web pages, VLEs)
• Signiﬁcant learning and teaching challenges can be addressed by
accessing resources across departments, schools, institutions
• Argumentation and critical thinking could beneﬁt from advanced
reasoning over a large scale of resources
• Could we adopt a bottom-up approach starting from linked data which
can be related to (layers of) ontologies later in the context of speciﬁc
➡ VALUE IN A LINKED DATA FIELD ACROSS HE
breadth vs. depth
• The value of semantic technologies on a large scale needs to be
- In addition to the value of reasoning using ontologies
• Mapping a critical volume of linked data to expressive ontologies
can be promising
• Encouragement for community-agreed ontologies can be more
effective and ﬂexible
• Expressive semantics to enable pedagogy-aware applications over a
large volume of linked data can be meaningful in a Web 2.0+ world
a roadmap of sem tech adoption
• Stage 1
‣ Exposing internal repositories as linked data, Performance optimised triple stores
‣ Searching across repositories, matching students, teachers, curricula, research interests
• Stage 2
‣ Advanced searching and matching, argumentation and critical thinking applications
‣ Mapping linked data to application-wide or community-wide agreed ontologies
• Stage 3
‣ Collaborative semantic enrichment of linked data relevant to communities
‣ Pedagogy aware reasoning applications and services
the network effect
• HE institutions exposing relational databases, VLE material, Web pages as
‣ Relevant technologies: RDF, RDFa, VLE plugins
‣ Starting from information already available in (X)HTML!
• Applications that use exposed linked data across institutions
‣ Curriculum design or alignment
‣ Inline recommendation of resources or people
‣ Applications that map linked data to community agreed or idiosyncratic
ontologies for critical thinking, argumentation, ...
ﬁrst 100 weeks?
• HE institutional, learning & teaching challenges:
• What information has been/can be used to address them?
• Where can this information be found in institutional repositories?
• How does this information differ across disciplines?
• Can this information be aligned across the HE sector?
• Can this information be related to other open repositories?
• What are the institutional barriers?
ﬁrst 100 weeks?
• Requirements of a linked data infrastructure for HE?
• Affordances of linked data repositories for the scale of HE?
• Query language requirements to support relevant queries?
• Interoperability with institutional repositories
• databases, VLEs, web pages, XML documents, ...
• Authenticity and integrity of HE resources?
• Conﬁdentiality requirements? Anonymisation?
• Interoperability with existing open repositories, knowledge models
Thanassis Tiropanis - firstname.lastname@example.org
The project: semtech-survey.ecs.soton.ac.uk
The survey: www.semtech.ecs.soton.ac.uk
The workshop: www.semhe.org