• Share
  • Email
  • Embed
  • Like
  • Save
  • Private Content
Personalization & Adaptivity
 

Personalization & Adaptivity

on

  • 2,236 views

Presented to IMS Global Conference, San Diego, 2013

Presented to IMS Global Conference, San Diego, 2013

Statistics

Views

Total Views
2,236
Views on SlideShare
2,145
Embed Views
91

Actions

Likes
3
Downloads
31
Comments
1

2 Embeds 91

https://twitter.com 63
http://www.scoop.it 28

Accessibility

Categories

Upload Details

Uploaded via as Microsoft PowerPoint

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel

11 of 1 previous next

  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment
  • Baker, R. S. J.d., & Yacef, K. (2009). The state of educational data mining in 2009: A review and future visions. Journal of Educational Data Mining, 1(1). http://www.educationaldatamining.org/JEDM/images/articles/vol1/issue1/JEDMVol1Issue1_BakerYacef.pdf
  • Bienkowski, M., Feng, M., & Means, B. (2012). Enhancing teaching and learning through educational data mining and learning analytics. U.S. Department of Education. Retrieved on March 10, 2013 from http://www.ed.gov/edblogs/technology/files/2012/03/edm-la-brief.pdf
  • State of Human Capital, 2012, http://www.mckinsey.com/~/media/McKinsey/dotcom/client_service/Organization/PDFs/State_of_human_capital_2012.ashx
  • https://www.mckinseyquarterly.com/Economic_Studies/Productivity_Performance/Preparing_for_a_new_era_of_knowledge_work_3034
  • http://www.census.gov/prod/2013pubs/acsbr11-14.pdf
  • Rich, E. (1979). User modeling via stereotypes. Cognitive Science 3, 329-354.Fischer, G. (2001). User Modeling in Human-Computer Interactions. User Modeling and User-Adapted Interaction, 11, 65-86.
  • http://scimaps.org/maps/map/mapping_the_evolutio_81/
  • http://onlinelibrary.wiley.com/doi/10.1002/tea.3660271003/abstract (Novak)Brusilovsky, P. (2001). Adaptive hypermedia: From intelligent tutoring systems to web-based education. User Modeling and User-Adapted Interaction, 11(1-2), 87-110.
  • Anderson, J. R., Corbett, A. T., Koedinger, K. R., & Pelletier, R. (1995). Cognitive tutors: Lessons learned. The Journal of the Learning Sciences, 4(2), 167-207.Burns, H. L. (1989). Foundations of intelligent tutoring systems: An introduction. In Richardson, J. J., & Polson, M. C. (Eds.), Proceedings of the Air Force Forum for Intelligent Tutoring Systems. http://www.dtic.mil/cgi-bin/GetTRDoc?AD=ADA207096#page=16
  • http://walker.wi.gov/Images/News/6.19.12%20UW%20Flexible%20Degree%20Proposal%20Packet.pdf

Personalization & Adaptivity Personalization & Adaptivity Presentation Transcript

  • Turning Data into PersonalizedStudent ExperiencesGeorge Siemens, PhDMay 14, 2013Presented toIMS Global
  • Technique:Baker and Yacef (2009) five primary areas ofanalysis:- Prediction- Clustering- Relationship mining- Distillation of data for human judgment- Discovery with models
  • Application: Bienkowski, Feng, and Means(2012)five areas of LA/EDM application:- Modeling user knowledge, behavior, and experience- Creating profiles of users- Modeling knowledge domains- Trend analysis- Personalization and adaptation
  • LA approach ExampleTechniquesModeling Attention metadataLearner modelingBehavior modelingUser profile developmentRelationship Mining Discourse analysisSentiment analysisA/B TestingNeural networksKnowledge Domain Modeling Natural language processingOntology developmentAssessment (matching user knowledge withknowledge domain)Siemens 2013: Adapted from Bienkowski et al, 2012, Baker & Yacef, 2009, Baker & Siemens 2013
  • LA approach ExampleApplicationsTrend Analysis and Prediction Early warning, risk identificationMeasuring impact of interventionsChanges in learner behavior, course discussions,identification of error propagationPersonalization/AdaptivelearningRecommendations: content and socialconnectionsAdaptive content provision to learnersAttention metadataStructural analysis Social network analysisLatent semantic analysisInformation flow analysisSiemens 2013: Adapted from Bienkowski et al, 2012, Baker & Yacef, 2009, Baker & Siemens 2013
  • Context
  • The Conference Board& McKinsey & Co
  • McKinsey Quarterly, 2012
  • Increasing diversityof student profilesThe U.S. is now in a position when less thanhalf of students could be considered fulltimestudents. In other words, students who canattend campus five days a week nine-to-five,are now a minority.(Bates, 2013)
  • Increasingly: learning acrosstraditional boundaries(i.e. work, outside of classroom, hobby)
  • Ok, on to adaptivity, personalization
  • LA approach ExampleApplicationsTrend Analysis andPredictionEarly warning, risk identificationMeasuring impact of interventionsChanges in learner behavior, course discussions,identification of error propagationPersonalization/AdaptivelearningRecommendations: content and socialconnectionsAdaptive content provision to learnersAttention metadataStructural analysis Social network analysisLatent semantic analysisInformation flow analysisSiemens 2013: Adapted from Bienkowski et al, 2012, Baker & Yacef, 2009, Baker & Siemens 2013Personalization as the holy grail of learning
  • (btw – this isn’t new)Rich, 1979All those CMU folksFischer, 2001
  • How does it work?
  • First, a knowledge domain ismapped
  • http://www.plosone.org/article/info:doi/10.1371/journal.pone.0004803
  • http://drunks-and-lampposts.com/2012/06/13/graphing-the-history-of-philosophy/
  • http://drunks-and-lampposts.com/2012/06/13/graphing-the-history-of-philosophy/
  • http://linkeddata.org/
  • (again, not new)Novak, 1990 (concept mapping)Semantic web: Berners-Lee, Hendler,Lassila, 2001Brusilovsky, 2001
  • Next, the learner ismodeled/profiled
  • Cognitive stylesCognitive modelsLearning preferences (by various criteria)Tutors (cognitive, intelligent)
  • (Also, not new)Anderson, Corbett, Koedinger, Pelletier,1995That shady learning styles literatureBurns, 1989
  • Knowledge domain +learner profile/knowledge +?= Personalization!The ? varies: from algorithms to pixie dust to chicken bones
  • State of Wisconsin, 2012
  • State of Wisconsin, 2012
  • So, what about creativeprocesses?AI/ML/analytics aren’t usefulhere, are they?
  • “We’ve been interested in pushingcomputing to a new direction, computationalcreativity. We’re trying to draw on datasets, not just to make inferences about theworld, but to create new things you’venever seen”Lav Varshney on Watsonhttp://www.fastcodesign.com/1672444/try-a-recipe-devised-by-ibms-supercomputer-chef
  • “An Ecuadorian strawberry dessert algorithmicallymaximized for pleasantness”http://www.fastcodesign.com/1672444/try-a-recipe-devised-by-ibms-supercomputer-chef
  • “For as much as $20,000 per script…a team of analystscompare the story structure and genre of a draft script withthose of released movies, looking for clues to box-officesuccess.”
  • The need to sensemake
  • Sensemaking“Sensemaking is a motivated, continuouseffort to understand connections . . . in orderto anticipate their trajectories and acteffectively”(Klein et al. 2006)
  • or“Sensemaking is about labeling andcategorizing to stabilize the streaming ofexperience”(Weick et al. 2005: 411)
  • We socially sensemake throughstories, narratives, knowledgeexchange, discourse
  • We turn to technical approacheswhen the data exceeds ourcapacity to create social discoursearound itBut, in fairness, once we technically sensemake, we turn to narrative to share
  • Adaptivity/Personalizationaddresses these quadrants
  • The future of work is inthese quadrants
  • LA interoperability
  • Open Learning Analytics
  • 47
  • Twitter/Gmail:gsiemens