Videolectures for ocwc2010


Published on

Published in: Technology, Education
1 Comment
  • ... im always surprised when i see how others are using my design and my personal pictures without my permission. It would be fair to ask or at least mention source.
    Are you sure you want to  Yes  No
    Your message goes here
  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Videolectures for ocwc2010

  1. 1. exchange ideas / share knowledge<br />VIDEOLECTURES.NET<br />
  2. 2. Outline of the talk<br />About and K4A<br />Technical solutions in preparation<br />Towards the content personalisation<br />Automatic Transcriptions<br />Enhanced Recommender Services<br />Visitors analytics<br />OCWC on<br />
  3. 3. Jozef Stefan Institute Department of Knowledge Technologies @ Center for Knowledge Transfer<br />Jozef Stefan Institute (JSI) is the leading Slovene research institution for natural sciences (800+ people) in the areas of computer science, physics, chemistry<br />Document-Atlas<br />Department of Knowledge Technologies have around 60 people working in various areas of artificial intelligence (machine learning, data mining, semantic technologies, computational linguistics, decision support)<br />Spinoff-s: Cyc-Europe, Quintelligence, LiveNetLife, Temida, XLab<br />Selection of Portals and Products: <br /><ul><li>Text-Garden (
  4. 4. Enrycher(
  5. 5. VideoLectures.NET (
  6. 6. IST-World (
  7. 7. Project Intelligence (
  8. 8. Search-Point (
  9. 9. OntoGen(
  10. 10. Document-Atlas (
  11. 11. AnswerArt(</li></ul>Semantic-Graphs<br />VideoLectures.NET<br />Selection of FP6 & FP7 Projects (Integrated Projects and Networks of Excellence only):<br />FP7 IP ACTIVE – Enabling the Knowledge Powered Enterprise<br />FP7 IP COIN – COllaboration and INteroperability for networked enterprises<br />FP7 IP EURIDICE – Inter-Disciplinary Research on Intelligent Cargo for Efficient, Safe and Environment-friendly Logistics<br />FP7 NoE PASCAL2 – Pattern Analysis, Statistical Modeling and Computational Learning <br />FP7 NoE T4ME – Machine Translation & Multilingual Information Retrieval<br />FP6 IP NeOn– Lifecycle Support for Networked Ontologies<br />FP6 IP ECOLEAD – European Collaborative Networked Organizations Leadership Initiative<br />FP6 IP SEKT – Semantically-Enabled Knowledge Technologies<br />
  12. 12. Videolectures: Basic facts<br />10000 videolectures - CC<br />10000 unique visitors per day<br />Recorded events 2009: 70, 2868 videos<br />Shared business models:<br />Research projects <br />Events<br />Academic institutions<br />Baseline funds<br />In-house developed services with strong support in research in semantics<br />JSI infrastructure, 5 permanent, 10-15 part time<br />Goal: Contributing to a global higher ed change by offering open access to high quality scientific material<br />
  13. 13. International dimension<br />European research supported by the European Commission (from 3M to 10M Euro scale RTD projects)<br />International institutions: EC, CEEMAN , CERN , Cluster Network , EFMD, IPSA , CLSP, MIT, UC Irvine , Yale, Stanford, TEDx, CMU, University of Ljubljana, Slovenian public research agency…<br />Active participation in: Opencast, OCWC, EuroCRIS<br />Knowledge4All foundation<br />
  14. 14. K4A<br />Originates from Pascal NoE<br />Knowledge and content exchange network<br />Inspired and lead by most active institutions and organisations around the world from the area of free and open scientific content<br />Effective and pragmatic<br />Global impact<br />Distributed, networked, bottom –up governance<br />Funds , joint projects<br />Using existing University networks and resources<br />Distinctive element: all content to be scientifically approved<br />
  15. 15. K4A - Five pillars of activity<br />Infrastructure: ICT Matterhorn - Interoperability, Channels, Semantics<br />Science: Journal and conferences<br />Online scientific video journal to global university<br />Education: courses and content<br />Quality assurance – peer reviewed content<br />Research: <br />facilitating the systems, accessing the content, enabling interaction<br />IPRs, multilinguality, standards<br />Business models (added value models)<br />Other continent connections: case study in engagement and interaction<br />
  16. 16. World Summit Award 09<br />World Summit award 09 <br />“With this, “Videolectures.Net” has approximately outrun 20.000 other products and projects from 157 countries participating in the 4th edition of the WSA, the United Nations based contest for e-content and creativity in the Information Society”.<br />
  17. 17. Technology stack<br />5 servers serving 20 TB of data<br />700,000 unique files<br />300,000 web requests daily (90,000 dynamic) <br />
  18. 18. Technologies and Research<br />Deep Semantics & Reasoning (Cyc)<br />Light-Weight Semantic Technologies<br />(OntoGen, OntoBridge)<br />Decision Support (DEX)<br />Social Computing/Web2.0 (LiveNetLife)<br />Computational Linguistics <br />(Enrycher, AnswerArt)<br />Complex Data Visualization <br />(DocAtlas, NewsExplorer, SearchPoint)<br />Graph/Social Network Analysis <br />(GraphGarden/SNAP, IST-World, FPIntelligence)<br />Data/Web/Text/Stream-Mining<br />(TextGarden Suite of tools)<br />Statistical Machine Learning<br />
  19. 19. Personalisation<br />Log files<br />Content mining<br />Adaptation<br />Modeling<br />(Needs and preferences)<br />
  20. 20. Towards personalisation @<br />
  21. 21. User profiling service(Qminer)<br />Ver1 – identifying segments: developed for NYT, Bloomberg <br />Ver2 – individual profiling: web service for<br />Analysing user logs and the content being accessed<br />Textual description – need for transcripts<br />Contextualisation – need for enriched content<br />Deep analytics<br />Modeling user behavior<br />Detecting SIGs – marketing groups, investors,…<br />Predicting and simulating user’s<br />Detecting trends in visits<br />Personalising content and methods<br />…<br />
  22. 22. User profiling – identifying segments<br />QMiner<br />System/services<br />Log <br />files<br />User profiles <br />Search fields<br />Search field values<br />Add state<br />Non-persistent Query<br />Get state<br />Get states<br />Update<br />Rename state<br />Delete state<br />Change Index<br />Exit<br />Videos articles<br />Editors<br />Advertisers<br />Authors<br />
  23. 23. Recommendation service(Recommender)<br />Ver1: Developed and tested for<br />Ver2: Operating at also for textual documents<br />Each video is scored from three directions:<br />Collaborative filtering<br />Category – VL taxonomy and improved SVM module working on optimized categories<br />Content – matching video against the user group’s history using all the enriched features<br />All three scores are combined into final score using weights estimated from the collected training data<br />Demonstration<br />
  24. 24. Content enrichment(enrycher)<br />Providing wider context to the document<br />… needed for efficient content mining and modeling<br />A set of Web services (<br />Enriching a document with annotations presenting:<br />Extracted known concepts to the machine<br />Generated most descriptive sentences and dynamic abstracts<br />Semantic graph<br />Descriptions with existing ontologies<br />Links to the external sources (wikipedia, dmoz, dbpedia, openlink data)<br />Demonstration<br />
  25. 25. Transcription service(Transcriptor)<br />Prototype service with automatic rapid vocabulary training of the speech recognition engine using:<br />Lecture description<br />Slides information<br />Videolectures taxonomy<br />Enriched complementary content<br />Used for:<br />Transcription<br />Speech indexing<br />Video content search<br />Demonstration<br />
  26. 26. OCWC on videolectures.NET<br />Videolectures.NET offers to organisations:<br />Low cost service and channel<br />Unlimited video preservation and fixed urls<br />Organisation, project and personal videography pages<br />Access to the back-office editorial and tools<br />Many innovative viewing and content management features <br />Sustainable innovation through research projects<br />Demonstration<br />
  27. 27. Supporting OCWC<br />Video and courses content distribution through<br />User modeling and analytics <br />… on a distributed network of OCWC sites<br />… common access to the analytics services<br />Opening existing services for independent use<br />… transcription, categorisation, classification, content enrichment<br />OCWC website on<br />… crawling, enriching, structuring, categorisingdistributed materials<br />… common curriculum support<br />
  28. 28. – head of Center for knowledge transfer at JSI<br /> – head of service<br /> – main editor at<br /> – head of the KT research group at JSI<br />John Shawe -Taylor ( – K4A director<br />Colin de la Higuera ( – K4A director<br />Enrycher:<br />Recommender:<br />Contextual search:<br />
  29. 29. Support slides<br />
  30. 30. A movement/competition …<br />
  31. 31. Competitive advantage<br />Access to lecture rooms and the three most active communities<br />Videos + slides + comments<br />Viewing features<br />Semantically enriched functionalities<br />Curriculum building and management support<br />Efficient back-office<br />Low cost and efficient service from recording to hosting<br />
  32. 32. Answering to challenges?<br />OpenCourseWare<br />MIT + >140 Universities<br />Curriculum, standards, quality of training<br />OpenCast<br />Berkeley, ETH + 40 top World Universities<br />OS for video recording at Universities<br />VL as CDCs<br />Knowledge4All foundation<br />Open CDN<br />Videolectures + JSI team<br />Using University Internet links and servers<br />
  33. 33. K4A founders<br />Europe – Pascal2 Network of Excellence:<br />University College London<br />Jozef Stefan Institute<br />University of Bristol <br />XEROX Research Centre Europe<br />ETH Zurich<br />CERN <br />US:<br />Berkeley + Opencast community<br />MIT + OCW consortium<br />Asia<br />Korea University + Network of South Korean Universities<br />Africa<br />Voices of Africa, Kenya + East Africa Universities <br />Kofi Annan Center for ICT and Development, Ghana + West Africa Universities<br />
  34. 34. K4A - reach<br />
  35. 35. Current development<br />OpenCDN – OSS/Collaborative Content Distribution Network<br />Automatic capturing, enriching, and synchronisation<br />Deep semantic search through videos<br />Accessibility, multilinguality<br />Knowledge extraction<br />Speech Indexing, Text Mining, Video mining, <br />Automatic ontology construction, <br />User Tracking and Profiling.<br />
  36. 36. SCOPE proposal<br />
  37. 37.
  38. 38. Visitors<br />
  39. 39. Knowledge 4 All<br />
  40. 40. Expressed interest<br />Internet Society Central America - Mexico<br />Individual organisations: Trento, ULJ, Zagreb, Southampton, CNRS, VTT, Max Planck, TU Graz, TUB, Oxford, Carlos III de Madrid, UVA,…<br />Commercial organisations: Springer Verlag, Elsevier Science<br />Governmental bodies: Slovenia, European Commission<br />
  41. 41. Development<br />Research<br />Added value (business) models<br />Emerging organisation models <br />Innovative tools <br />Operative<br />Methods (individual, collaborative, business)<br />Didactics, methodics, pedagogical models<br />Systems, standards, interoperability <br />Free, open access, high quality, scientific content <br />
  42. 42. Projects<br />In preparation:<br />AI Research institute for West Africa: implications for infrastructure, summer schools, course definition, interaction software, etc.<br />Education kiosks in Africa<br />Journal SCI registration – also in discussion with Springer about possible publication<br />Virtual conference<br />Virtual university<br />Web 2.5 for learning: support for discussion groups, research communities<br />
  43. 43. Long-term options<br />Innovation tube – industry/business use<br />Virtual universities and virtual programmes<br />Bottom-up, distributed, self-organised, <br />Authoring services<br />Support content enrichment for the content creators<br />Services:<br />On-the-fly personalisation and recommendation<br />Video scene recognition, automatic annotation and categorisation<br />Semantic and multilingual search<br />Accessibility, Internationalization (subtitles, transcripts)<br />Advanced presentation services with direct user involvement<br />Textual, graphical, video (audio) content integration services and enrichment<br />