Week 2 of the Learning Analytics (LAK11) course focuses on the emerging field of educational data mining (EDM). Data mining tries to make senses of huge amounts of data. These can be in tabular...
My comment on this was: If you data mined students responses to why they did such and such, would that not address a “why” question? In work I’ve been doing with Jody Clarke-Midura, the data being mined are the paths students have taken to arrive at a scientific conclusion. Patterns of activity found for subgroups in the data appear to us to be signs of both “how” and “why”. The paths might indicate how a certain group of students navigated some decisions and then why that led them to a particular conclusion. Group membership is then one of the predictive elements of the analysis; if someone can be shown to be a member of a certain group, then the probability of that student doing certain things and claiming certain things is increased.
I guess I should divide my scoops a bit - into analytics versus deeper learning...but I really hope these two can be integrated to personalize learning. So can we send a Hadoop team to Skytree to solve a university's challenge?
An overview of the Draft Issue Brief prepared by SRI International for the US Department of Education on Educational Data Mining and Learning Analytics (Educational Data Mining/Learning Analytics issue brief overview by @mariebienkowski on @slideshare...
In this post I’ve included the key developments of this past week that will keep readers in-the-know on education news. Another new MOOC platform, NovoEd launched by Stanford this week offers challenging courses and takes a unique approach to team projects and peer grading, and the machine grading of essays—the debate continues and is an issue that prevents one school from joining edX. Also, I’ll introduce a new tool that bring interactivity to online learning.
New Ed-Tech Tool to Support Interaction in Online Courses
This platform looks like its worthy of investigating further – as it provides easy way to build interactive content into online courses: “Smart Sparrow is an Australian ed-tech start-up pioneering adaptive and personalized learning technology. It was founded by Dr Dror Ben Naim who led a research group in the field of Intelligent Tutoring Systems and Educational Data Mining at the University of New South Wales in Sydney resulting in the development of the Adaptive e-Learning Platform”.
"In the fall of 2011, Stanford University offered three of its engineering courses—Artificial Intelligence, Machine Learning, and Introduction to Databases—for free online... As Sebastian Thrun, the director of Stanford’s Artificial Intelligence Laboratory, tells the story, he assumed just a handful of people would enroll in his graduate-level AI class. Instead, more than 160,000 students registered. A massive number. That’s when the enormous hype began about massive open online courses, better known as “MOOCs.” Since then, Thrun and his fellow lab professors Daphne Koller and Andrew Ng have founded education organizations that offer free online classes."
I've been dipping my toes into the two cultures (love the ref to the C.P. Snow term) and really appreciate the added details in this set of blog links and discussion points, comparing stats and machine learning - also, do follow the link on data mining, which is different from machine learning...maybe we need a "3 cultures" article!
What do brilliant data scientists, roller derby enthusiasts, and Prince William lookalikes have in common? You can find all three at Kaggle, a company whose mission is to bring together the top data scientists in the world to solve complex problems.
Need an automated essay scoring tool? a predcitive analytic solution to retention in college programs? a game-based learning algorithm? How about using a crowd-sourcing method...at Kaggle.
Interesting concept that envisions intelligence as an energy-organizational issue so it applies thermodynamics principles to decisions. The issue of organization arising from chaos does involve local organization against the global surrounding entropy - hmmm.
This is an explanation and demo of software I wrote that learns how to play a Nintendo Entertainment System game and then automatically plays it. This is rea...
One of the pictures of the future of learning? This is intriguing - seeming unrelated but I sense there might be a connection from his research over to the problem of finding expert pathways in any digital learning space.
Without scanning the whole article, what do you think we can do to dive deeper into use patterns? That is, how can we go deeper than what is mentioned in this quote:
"They found that people take classes or stop for different reasons, and therefore referring globally to "dropouts" makes no sense in the online context. They identified four groups of participants: those who completed most assignments, those who audited, those who gradually disengaged and those who sporadically sampled. (Most students who sign up never actually show up, making their inclusion in the data problematic.) The point of all this is not simply to record who is doing what but to "provide educators, instructional designers and platform developers with insights for designing effective and potentially adaptive learning environments that best meet the needs of MOOC participants," the researchers wrote."
New York Times New artificial-intelligence system grades essays at college level Denver Post Imagine taking a college exam and, instead of handing in a blue book and getting a grade from a professor weeks later, clicking the "send" button when you...