Artificial intelligence (ai) personalization and learning

Launching and leveraging adaptive learning: research, efficacy, and results presentation at GlobalMindset Sydney, Australia

1. Artificial Intelligence (AI), Personalization and Learning: Will it Work? Karen Vignare, PhD, MBA CEO & Founder, KV Consulting 17th Global Mindset conference on Emerging Trends in Learning & Working Sydney - 03rd August 2016

2. Agenda About me Separating fact from fiction in AI What do we know about adaptive learning technologies? Piloting adaptive learning tools The research results Lessons Learned

3. Karen Vignare, PhD, MBA 20 years experience online, emerging technologies in learning Started online programs, worked with international universities Managed blended learning, MOOCs, and adaptive learning projects Extensive research portfolio More about me at LinkedIn, https ://www.linkedin.com/in/karen-vignare

4. AI for Learning: Definitions The central problems (or goals) of AI research include  reasoning ,  knowledge ,  planning ,  learning ,  natural language processing  (communication),  …….Approaches include  statistical methods ,  computational intelligence ,  soft computing  (e.g.  machine learning ), and  traditional symbolic AI . Many tools are used in AI, including versions of  search and mathematical optimization ,  logic ,  methods based on probability and economics . The AI field draws upon  computer science ,  mathematics ,  psychology ,  linguistics ,  philosophy ,  neuroscience  and  artificial psychology . Source: Wikipedia

5. New technologies for learning Personalization does not have to mean technology but it most cases organizations are leveraging technologies (planning, advising, relationship management, courses, analytics and tracking to personalize) Adaptive learning and digital courseware technologies focus more on personalizing or adapting the course content and assessments to the individual students The tools rely on algorithms that direct learners based on the content and the paths they choose For implementers/instructors reading the analytics and using that information is a critical new role

6. What does the research say? Connected to progress made in Computer Aided Instruction (circa 1980s) and research pointed to increased effectiveness Meta-analyses showed promise and effectiveness but disagreement on approach and value remained Internet slowed gains as more people explored constructivism, connectivism Current online tools seem to be at same effectiveness as three decades ago… Costs could be lower, algorithms more powerful, but design and computer modeling are still debatable

7. Current State: Most Rely on LMS Technology LMS is still key to supporting online learning in higher education and other tools are slowly being tested and brought in

8. The Challenge & The Opportunity The future is Now : Leveraging digital tools & resources to improve instruction & outcomes

9. Another View of Digital Learning Environment

10. Current State of Adaptive Learning Success Next Generation Digital Learning Environment Courseware in Context Framework Reported pilots Interoperability and Integration Latest version compliance LTI Most report easy interoperability Personalization Critical within system No clear standard match so reports are more on reported outcome success Analytics, Advising, and Learning Assessment Caliper and LTI Needs improvement as data exchange is not widely used (yet) Collaboration Currently not as important Accessibility and Universal Design Critical to scale Major concerns (not just a vendor issue)

11. Gartner Education Hype Cycle 2015

12. Some of the Companies in the Space Source: Tyton Partners

13. Source: https://www.edsurge.com/higher-ed/special-projects/dl/orientation

14. Source: http://coursewareincontext.org/

15. Leaders/Managers are Underprepared The business is changing—accountability, stakeholders are all demanding more Outdated systems and processes Very little training for administrators Many became leaders decades ago and are not capable in current business operations, technologies and processes Higher education does not always attract the best “business” people but does attract “researchers/scholars”

16. Piloting Adaptive Learning At last job, used six different adaptive tools Vendors range from publishers, platform tools, to university created ones Increased technology sophistication Focus is on content variety, assessment breadth, and immediacy As technology substitutions for online learning, mixed results Within tool correlations are very promising

17. Content Map for Types of statistical studies Source: https://www.khanacademy.org/exercisedashboard

18. Lessons Learned The future of adaptive learning in higher education depends on the commitment level of universities Preparation ...really take the time to understand the power of the tool, take a class Skill up faculty, instructional staff and technology team Pick your course(s) based on solving problems (is content difficult, would more student practice help, does immediacy help students Build content maps and you will need more content than you currently have (unless you use pre-packaged) Learn to use the dashboard Iterate probably at least three times....

19. Questions Karen Vignare, PhD, MBA [email protected] 240.462.2160