0
Linked Data and Open Educational Resources -      towards a symbiotic Relationship                     Stefan Dietze      ...
IntroductionResearch areas                                                                                                ...
RecSys for TEL; Learning AnalyticsCollab./content- based RecSys                                            Open Educationa...
TEL environments & recommender systems dependent on                    availability of data                        “In the...
Educational Web dataState  Vast Open Educational Resource (OER) metadata collections  (e.g. OpenCourseware, OpenLearn, Mer...
Web-scale exploration of educationalresources and data ?      RecSys for TEL; Learning AnalyticsCollab./content- based Rec...
RecSys for TEL; Learning Analytics         Educational Linked Data                                                        ...
(Linked) Open Data                                                                (c) Paul Miller                     tele...
(Linked) Open Data  Linked (Open) Data – “Semantic Web done right”     Vision: well connected graph of open Web data     W...
(Linked) Open Data for EducationDatasets which might enhance (informal) learning  Publications & literature: ACM, PubMed, ...
(Linked) Open Data for EducationDatasets which might enhance (informal) learning  Publications & literature: ACM, PubMed, ...
(Linked) Open Data for Education  Applications of educational LOD  (eg from past projects & LILE2012)    Web-search of edu...
Web-scale TEL data exploitation                                                      Applications/tools exploiting TEL Web...
Web-scale TEL data exploitation                                                      Applications/tools exploiting TEL Web...
mEducator                   => http://www.meducator.net    EC-funded eContentPlus Best Practice Network (BPN) ,    May 200...
Challenges & approach1. Improving OER metadata interoperability and Web-wide search by applying LOD principles…2. …WHILE e...
Challenges & approach1. Improving OER metadata interoperability and Web-wide search by applying LOD principles…2. …WHILE e...
Application context: biomedical education                              Metamorphosis+         Tailored (L)CMS plugins     ...
Approach: educational service integration SmartLink: Linked Data registry of (educational) datasets / stores and their API...
Approach: educational data integration Issue: often poorly structured metadata, free-text and proprietary taxonomies Goal:...
(1) Enrichment: automated via DBpedia & Freebase                                                   Semi-structured RDF    ...
(1) Enrichment: automated via DBpedia & Freebase                                                   Semi-structured RDF    ...
(1) Enrichment: automated via DBpedia & Freebase                 ?                                                        ...
(1) Enrichment: automated via DBpedia & Freebase                                                       NER & disambiguatio...
(1) Enrichment: semi-automated   Example: OER annotation in MetaMorphosis+                                                ...
(1) Enrichment: semi-automated Access to 324 ontologies                           1. User-specified term during and over 5...
(2) Structural clustering of related resourcesNumber of resources per DBpedia reference/enrichment (subject) in mEducator ...
(2) Structural clustering of related resourcesNumber of resources per DBpedia reference/enrichment (subject) in mEducator ...
(2) Clustering (similarity-based, linguistic)Vector-based similarity computation based on:1) Data indexing => Doc-Term Mat...
Exploratory search enabled via clusteringExample: search results of OER in MetaMorphosis+                              Met...
Exploratory search enabled via clusteringExample: search results of OER in MetaMorphosis+                       Metamorpho...
Data so far: SmartLink/mEducator in LOD cloud http://ckan.net/package/smartlink > 2000 triples so far > 300 links to iServ...
Web-scale TEL data exploitation                                                      Applications/tools exploiting TEL Web...
Educational Web data: open issuesMotivation  Quality and quantity of (educational) Web data constantly improving  Exploita...
LinkedUp in a nutshell                                                        Applications and tools                      ...
LinkedUp consortium Web data integration & TEL & Open Data dissemination                                                  ...
LinkedUp   Exploitation, dissemination, sustainability   Persistent “LinkedUp Network”(extensible community of industrial ...
LinkedUpNext stepsOngoing preparations to enable quickstart (1 November 2012)   Challenge design, community & clusters   C...
?RecSys for TEL; Learning Analytics         Educational Linked Data            RecSys                                     ...
Linked Data & TEL – a symbiotic relationship!                                    requires data  Improving in terms of scal...
Thank you!                                   http://purl.org/dietze                                 http://linkededucation...
Upcoming SlideShare
Loading in...5
×

Linked Data vs Open Educational Resources

1,456

Published on

Talk on Linked Data in Education held at 6th tele-TASK Symposium, Hasso Plattner Institute (HPI) Potsdam, Germany, 9 October 2012

Published in: Career
0 Comments
5 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total Views
1,456
On Slideshare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
49
Comments
0
Likes
5
Embeds 0
No embeds

No notes for slide

Transcript of "Linked Data vs Open Educational Resources"

  1. 1. Linked Data and Open Educational Resources - towards a symbiotic Relationship Stefan Dietze - 6th tele-TASK Symposium 2012 -
  2. 2. IntroductionResearch areas ⇒ http://www.l3s.de/ Semantic Web & Linked Data, data & knowledge ⇒ http://kmi.open.ac.uk/ integration (mapping, classification, interlinking) Application domains: education/TEL, Web archiving, …Projects & activities EU funded research projects: (Linked) Web data & education „Linked Learning“ and „LALD“ workshops (eg LILE2012@WWW2012) http://linkededucation.org & http://linkeduniversities.org More information: http://purl.org/dietze tele-TASK Symposium 2012 Stefan Dietze 2
  3. 3. RecSys for TEL; Learning AnalyticsCollab./content- based RecSys Open Educational Resources RecSys TEL
  4. 4. TEL environments & recommender systems dependent on availability of data “In the lab”: data for evaluation “In the wild”: real-world TEL applications Quantity, quality (e.g. accessibility, interoperability) and reusability of Web data (in particular about OER) is crucial RecSys for TEL; Learning AnalyticsCollab./content- based RecSys Open Educational Resources RecSys TEL
  5. 5. Educational Web dataState Vast Open Educational Resource (OER) metadata collections (e.g. OpenCourseware, OpenLearn, Merlot, ARIADNE) Usually exposed via APIs/services (c) Paul Miller Competing Web interfaces (e.g. SQI, OAI-PMH, SOAP, REST) Competing metadata standards (e.g. IEEE LOM, ADL SCORM, DC…) Competing exchange formats and serialisations (e.g. JSON, RDF, XML) Fragmented use of taxonomiesIssues Heterogeneity & lack of interoperability Lack of take-up tele-TASK Symposium 2012 Stefan Dietze 5
  6. 6. Web-scale exploration of educationalresources and data ? RecSys for TEL; Learning AnalyticsCollab./content- based RecSys Open Educational Resources RecSys TEL
  7. 7. RecSys for TEL; Learning Analytics Educational Linked Data Linked Open DataCollab./content- based RecSys Semantic Web Open Educational Resources RecSys Semantic Web TEL
  8. 8. (Linked) Open Data (c) Paul Miller tele-TASK Symposium 2012 Stefan Dietze 8
  9. 9. (Linked) Open Data Linked (Open) Data – “Semantic Web done right” Vision: well connected graph of open Web data W3C standards (RDF, SPARQL) to expose data, URIs to interlink datasets => vast cloud of interconnected datasets Crossing all sorts of domains Number ofDomain Triples % datasetsMedia 25 1,841,852,061 5.82 %Geographic 31 6,145,532,484 19.43 %Government 49 13,315,009,400 42.09 %Publications 87 2,950,720,693 9.33 %Cross-domain 41 4,184,635,715 13.23 %Life sciences 41 3,036,336,004 9.60 %User-generated 20 134,127,413 0.42 %content 295 31,634,213,770 Source: http://lod-cloud.net/state, September 2011 tele-TASK Symposium 2012 Stefan Dietze 9
  10. 10. (Linked) Open Data for EducationDatasets which might enhance (informal) learning Publications & literature: ACM, PubMed, DBLP (L3S), OpenLibrary Domain-specific knowledge & resources: Bioportal for Life Sciences, historic artefacts in Europeana, Geonames Cross-domain knowledge: DBpedia, Freebase, … Media resource metadata: BBC, Flickr, … tele-TASK Symposium 2012 Stefan Dietze 10
  11. 11. (Linked) Open Data for EducationDatasets which might enhance (informal) learning Publications & literature: ACM, PubMed, DBLP (L3S), OpenLibrary Domain-specific knowledge & resources: Bioportal for Life Sciences, historic artefacts in Europeana, Geonames Cross-domain knowledge: DBpedia, Freebase, … Media resource metadata: BBC, Flickr, …Explicitly educational datasets and schemas University Linked Data: eg The Open University UK, http://data.open.ac.uk, Southampton University, University of Munster (DE), http://education.data.gov.uk OER Linked Data: mEducator Linked ER (http://ckan.net/package/meducator), Open Learn LD Schemas: Learning Resource Metadata Initiative (LRMI, http://www.lrmi.net/), mEducator Educational Resources schema (http://purl.org/meducator/ns)=> see http://linkededucation.org & http://linkeduniversities.org tele-TASK Symposium 2012 Stefan Dietze 11
  12. 12. (Linked) Open Data for Education Applications of educational LOD (eg from past projects & LILE2012) Web-search of educational courses/OER („educational graph“) Game-based learning & automatic generation of assessment items from LOD Enrichment of learning resources (facilitating more exploratory learning approaches)….http://metamorphosis.med.duth.gr/ tele-TASK Symposium 2012 Stefan Dietze 12
  13. 13. Web-scale TEL data exploitation Applications/tools exploiting TEL Web data for recommendation/exploration: scalability, robustness licensing and legal issues Challenges Web-scale TEL data integration data quality (ambiguity, richness, …) data heterogeneity (semantic), data interlinking RecSys for TEL; Learning Analytics Educational Linked Data Linked Open DataCollab./content- based RecSys Semantic Web Open Educational Resources RecSys Semantic Web TEL
  14. 14. Web-scale TEL data exploitation Applications/tools exploiting TEL Web data for recommendation/exploration: scalability, robustness licensing and legal issues Web-scale TEL data integration data quality (ambiguity, richness, …) data heterogeneity (semantic), data interlinking RecSys for TEL; Learning Analytics Educational Linked Data Linked Open DataCollab./content- based RecSys Semantic Web Open Educational Resources RecSys Semantic Web TEL
  15. 15. mEducator => http://www.meducator.net EC-funded eContentPlus Best Practice Network (BPN) , May 2009 – May 2012 (3 years duration) 14 partners:+ tele-TASK Symposium 2012 Stefan Dietze 15
  16. 16. Challenges & approach1. Improving OER metadata interoperability and Web-wide search by applying LOD principles…2. …WHILE exploiting existing OER metadata and infrastructures Open Educational Resources ? Linked Data tele-TASK Symposium 2012 Stefan Dietze 16
  17. 17. Challenges & approach1. Improving OER metadata interoperability and Web-wide search by applying LOD principles…2. …WHILE exploiting existing OER metadata and infrastructures Data/services integration & retrieval/search APIs tele-TASK Symposium 2012 Stefan Dietze 17
  18. 18. Application context: biomedical education Metamorphosis+ Tailored (L)CMS plugins => http://metamorphosis.med.duth.gr/ => http://www.meducator3.net/ Data/services integration & retrieval/search APIs tele-TASK Symposium 2012 Stefan Dietze 18
  19. 19. Approach: educational service integration SmartLink: Linked Data registry of (educational) datasets / stores and their APIs Discovery and lifting of educational data out of heterogeneous repositories Transformation of heterogeneous data formats (XML, JSON...) and schemas (eg. IEEE LOM, Dublin Core) into RDF (pre-requisite for LOD compliancy) ⇒ http://ckan.net/package/smartlink & http://purl.org/smartlink Data/services integration & retrieval/search APIs tele-TASK Symposium 2012 Stefan Dietze 19
  20. 20. Approach: educational data integration Issue: often poorly structured metadata, free-text and proprietary taxonomies Goal: improvement of lifted (RDF) data with public LOD vocabularies; tighter interlinking to provide coherent and well-connected graph of educational data (across disparate stores) Approach: 1) Data enrichment (via DBpedia, Freebase, BioPortal) 2) Clustering (structural as well as linguistic) to identify correlating resources ⇒ http://linkededucation.org/meducator Data/services integration & retrieval/search APIs Linked Educational Resources tele-TASK Symposium 2012 Stefan Dietze 20
  21. 21. (1) Enrichment: automated via DBpedia & Freebase Semi-structured RDF description of ? educational resource tele-TASK Symposium 2012 Stefan Dietze 21
  22. 22. (1) Enrichment: automated via DBpedia & Freebase Semi-structured RDF description of educational resource ? tele-TASK Symposium 2012 Stefan Dietze 22
  23. 23. (1) Enrichment: automated via DBpedia & Freebase ? ? ? ? Stefan Dietze 18/09/12 23
  24. 24. (1) Enrichment: automated via DBpedia & Freebase NER & disambiguation, eg, via ! ! Stefan Dietze 18/09/12 24
  25. 25. (1) Enrichment: semi-automated Example: OER annotation in MetaMorphosis+ Metamorphosis+ http://metamorphosis.med.duth.gr/ tele-TASK Symposium 2012 Stefan Dietze 25
  26. 26. (1) Enrichment: semi-automated Access to 324 ontologies 1. User-specified term during and over 5 Mio entities learning resource annotation Metamorphosis+ http://bioportal.bioontology.org/ http://metamorphosis.med.duth.gr/ 2. Suggested Entities 3. Selected entities from BioPortal used to describe discipline, keywords of resource tele-TASK Symposium 2012 Stefan Dietze 26
  27. 27. (2) Structural clustering of related resourcesNumber of resources per DBpedia reference/enrichment (subject) in mEducator dataset Cervical_cancer 59 Screening 31 Cervical 29 Hpv 29 Oxygenation 26 DBpedia references used most frequently to describe the Childhood 22 „subject“ of particular educational resources differential_diagnosis 19 Knowledge 18 Learning 17 decision_making 16 Training 15 Lecture 15 Risk 15 hpv_infection 15 Fear 15 pap_smear 15 Abnormal 14 Ventilation 14 Ecg 14 tele-TASK Symposium 2012 Stefan Dietze 27
  28. 28. (2) Structural clustering of related resourcesNumber of resources per DBpedia reference/enrichment (subject) in mEducator dataset Cervical_cancer 59 Screening 31 Clustering of resources graph (blue nodes: resources, green nodes: enrichments) Cervical 29 Hpv 29 Oxygenation 26 Childhood 22 differential_diagnosis 19 Knowledge 18 Learning 17 decision_making 16 Training 15 Lecture 15 Risk 15 hpv_infection 15 Fear 15 pap_smear 15 Abnormal 14 Ventilation 14 Ecg 14 Cluster of educational resources relating to „cervical cancer“ subject tele-TASK Symposium 2012 Stefan Dietze 28
  29. 29. (2) Clustering (similarity-based, linguistic)Vector-based similarity computation based on:1) Data indexing => Doc-Term Matrix (term frequencies in given resource metadata)2) Creation of similarity matrices => similarity values between resources3) Clustering (based on similarity thresholds) tele-TASK Symposium 2012 Stefan Dietze 29
  30. 30. Exploratory search enabled via clusteringExample: search results of OER in MetaMorphosis+ Metamorphosis+ http://metamorphosis.med.duth.gr/ Educational resources retrieved based on particular user query tele-TASK Symposium 2012 Stefan Dietze 30
  31. 31. Exploratory search enabled via clusteringExample: search results of OER in MetaMorphosis+ Metamorphosis+ http://metamorphosis.med.duth.gr/ Related resources (ranked) tele-TASK Symposium 2012 Stefan Dietze 31
  32. 32. Data so far: SmartLink/mEducator in LOD cloud http://ckan.net/package/smartlink > 2000 triples so far > 300 links to iServe APIs used by several applications http://ckan.net/package/meducator > 35000 triples so far > 1000 links to DBpedia & Bioportal ontologies APIs used by 4 applications tele-TASK Symposium 2012 Stefan Dietze 32
  33. 33. Web-scale TEL data exploitation Applications/tools exploiting TEL Web data for recommendation/exploration: scalability, robustness licensing and legal issues Web-scale TEL data integration data quality (ambiguity, richness, …) data heterogeneity (semantic), data interlinking RecSys for TEL; Learning Analytics Educational Linked Data Linked Open DataCollab./content- based RecSys Semantic Web Open Educational Resources RecSys Semantic Web TEL
  34. 34. Educational Web data: open issuesMotivation Quality and quantity of (educational) Web data constantly improving Exploitation of Web data lacking scale and often limited to few, mostly isolated datasetsLinking Web Data for Education Project – Open Challenge in Web-scale Data Integration EC Support Action, start November 2012, coordinated by L3S => http://linkedup-project.eu Goals Push forward adoption of Web data/Linked Data in educational context Drive technological advancement of Web data integration technologies (applications, IR technologies, recommender systems) Approach Open data competition; open education as big data scenario tele-TASK Symposium 2012 Stefan Dietze 34
  35. 35. LinkedUp in a nutshell Applications and tools TEL environments andLinkedUp in a nutshelldata Web applications Linked Open data Data integration tools: Educational data & resources (30+ billion statements) storage, analytics, OER metadata LinkedUp General Web data mining, integration, Web submissi OpenLearn data on data (OAI-PMH feeds, mapping web metadata etc) OpenCourseware … Personal Ariadne data … iTunesU Stage 1- EU project results Initialisation Initialisation 3 stages of the LinkedUp competition LinkedUp Challenge Environment • Lowest requirements level for participation • Inital prototypes and mockups, use of data • LinkedUp Evaluation Framework testbed required Participation criteria • Methods and Test Cases Stage 2 • 10 to 20 projects are expected • LinkedUp Data Testbed • Competitor ranking list • Medium requirements level for participation • Working prototypes, minimum amount of data sources, clear target user group LinkedUp Support Actions Challenge Stage 3 • 5 to 10 projects are expected • Dissemination (events, training) • Data sharing initiatives …provides support: • Deployment in real-world use cases • Community building & clustering • Sustainable technologies, reaching out • Technology transfer Financial awards to critical amount of users, Stage 4 • Cashprice awards & consulting • 3 to 5 projects are expected Legal & technical E P S guidance T P F Data & use cases Network of supporting organisations I (see 3.2 Spreading excellence, exploiting results, disseminating knowledge) S E C B C O tele-TASK Symposium 2012 Stefan Dietze 35
  36. 36. LinkedUp consortium Web data integration & TEL & Open Data dissemination L3S Research Center, Leibniz University, DEElsevier, NL Leading institute in Web science & Leading scientific & educational publisher data technologies as well as Innovative research on the future of publishing & technology-enhanced learning extensive experience in data competitions Coordinator and leader of LinkedUpCELSTEC, The Open University, NL Challenge WP R&D institute in educational technologies and part of the largest distance university in the netherlandsThe Open Knowledge Foundation, UK Not-for profit organisation to promote open knowledge and data; global network Host of key events (OKCon) and platforms (eg CKAN)KMI, The Open University, UK Leading R&D institute in areas related to LinkedUp World’s largest distance university (over 200.000 students)Exact Learning Solutions, IT SME in educational technologies and services with long-standing experience in (EC-funded) R&D projects tele-TASK Symposium 2012 Stefan Dietze Stefan Dietze 18/09/12 36
  37. 37. LinkedUp Exploitation, dissemination, sustainability Persistent “LinkedUp Network”(extensible community of industrial and academic institutions) Commonwealth of Learning, COL (CA) International Athabasca University (CA) (outside Europe) SURF NL (NL)Université Fribourg, eXascale Infolab Group (CH) Democritus University of Thrace (GR) AKSW, Universität Leipzig (DE) Aristotele University of Thessaloniki (GR) CNR Institute for Educational Technologies (IT) Clam Messina Service and Research Centre (IT) Eurix (IT) Ontology Engineering Group (OEG), UPM, (ESP) Stefan Dietze 18/09/12 37
  38. 38. LinkedUpNext stepsOngoing preparations to enable quickstart (1 November 2012) Challenge design, community & clusters Challenge kickoff: initial calls expected by February 2013 (http://www.linkedup-project.eu)Participate! As challenge participant Submission of innovative application/tool tackling one or more of the challenge goals LinkedUp offers: financial, technical and legal support As associated partner Participate as evaluation panelist, use case or data contributor & benefit from access to large network of organisations in Linked Data and TEL Take advantage of innovative research results (LinkedUp challenge submissions, evaluation framework) Promote your own data and tools tele-TASK Symposium 2012 Stefan Dietze 38
  39. 39. ?RecSys for TEL; Learning Analytics Educational Linked Data RecSys Semantic Web TEL
  40. 40. Linked Data & TEL – a symbiotic relationship! requires data Improving in terms of scalability, Wealth of relevant data accuracy, performance etc available, improving in terms of quantity and quality Challenge: availability and accessibility of diverse, high- Challenge: exploration and quality, interoperable data recommendation in large-scale distributed dataRecSys for TEL; Learning Analytics Educational Linked Data requires scalable IR/RecSys mechanisms RecSys Semantic Web TEL
  41. 41. Thank you! http://purl.org/dietze http://linkededucation.org http://linkedup-project.eu Some upcoming events Knowledge Extraction and Consolidation from Social Media (KECSM2012), workshop at ISWC2012, http://blogs.ecs.soton.ac.uk/knowledgeextraction/ tele-TASK Symposium 2012 Stefan Dietze 41
  1. A particular slide catching your eye?

    Clipping is a handy way to collect important slides you want to go back to later.

×