8 Ways Machine Learning Will Improve

Education
By Tom Vander Ark on November 25, 2015 7:00 AM | No comments

It was the constant stories about self-driving cars that made machine learning the
breakthrough tech topic of 2015. Five years ago news stories talked about robots that could
do repetitive tasks but they said real complex tasks requiring seasoned judgement--like
driving a car-- were years away. Then cars started driving themselves. It turns out that
computers are getting smart faster than most predicted.

Machine learning , a subset of artificial intelligence, is an effort to program computers to

identify patterns in data to inform algorithms that can make data-driven predictions or
decisions. As we interact with computers, we're continuously teaching them what we are like.
The more data, the smarter the algorithms become.

Pedro Domingos, author of the The Master Algorithm , said machine learning is the new
switchboard for HigherEd. Machine learning is the new weapon attacking cancer, climate
change, and terrorism. It's the new infrastructure for everything.

In the spring of 2014 data privacy (and over-testing) concerns rose to the forefront of the US
K-12 dialog. By October more than 100 EdTech vendors had signed a data privacy pledge.

In 2015, our SmartParents series argued that data is key to personalized learning and that
parents should have access to student data and should be able to decide with whom to share
portions of that data--requiring policymakers to embrace personalization and privacy.

This year it became apparent that machine learning and other big data strategies are quietly
improving formal and informal learning in many ways:

1. Content analytics that organize and optimize content modules:

 Gooru , IBM Watson Content Analytics

2. Learning analytics that track student knowledge and recommend next steps:

 Adaptive learning systems: DreamBox, ALEKS, Reasoning Mind, Knewton

 Game-based learning: ST Math, Mangahigh

3. Dynamic scheduling matches students that need help with teachers that have time:

 NewClassrooms uses learning analytics to schedule personalized math learning

experiences.

4. Grading systems that assess and score student responses to assessments and computer
assignments at large scale, either automatically or via peer grading:
  Pearson's WriteToLearn and Turnitin's Lightside can score essays and detect
plagiarism.

5. Process intelligence tools analyze large amounts of structured and unstructured data,
visualize workflows and identifying new opportunities:

 BrightBytes Clarity reviews research and best practices, creates evidence-based

frameworks, and provides a strength gap analysis.
 Enterprise Resource Planning (ERP) systems like Jenzabar and IBM SPSS helps
HigherEd institutions predict enrollment, improve financial aid, boost retention, and
enhancing campus security.

6. Matching teachers and schools:

 MyEdMatch and TeacherMatch are eHarmony for schools.

7. Predictive analytics and data mining to learn from expertise to:

 Map patterns of expert teachers

 Improve learning , retention, and application.

8. Lots of back office stuff:

 EDULOG does school bus scheduling

 Evolution , DietMaster.

Learning will remain highly relational for most of us, but those relationships will increasingly
be informed by data. Students, parents and advisors will make more decisions about learning
pathways but those decisions will be nudged and guided by informed recommendations.

In the coming year, every faculty should discuss the coming impact of big data--and ask
students to do the same.