Stanford scientists put free text-analysis tool on the web
Andrew Myers, Stanford Engineering, March 28, 2014

As this prromotional article states, "Ever wondered whether a certain TV show had a slant in favor of a political candidate? Stanford computer scientists have created a website that gives anyone who can cut and paste the ability to answer such questions, systematically and for free. The website is known as etcML, short for Easy Text Classification with Machine Learning." The story says two things to me. First, learning analytics is becoming a commodity, which will manifest itself as a service that other applications can access. Second, if you're developing something and want to develop a market for it, the way to be successful is to create a publicly-accessible protototype and then promote it - I know lots of people have been working on emotional and sentiment analysis, but the world beats a path to the door of the people who actually show how it's done. Via Geoffrey Rockwell.

[Link] [Comment]

Mapping Twitter Topic Networks: From Polarized Crowds to Community Clusters
Marc A. Smith, Lee Rainie, Ben Shneiderman, Itai Himelboim, Pew Research Internet Project, March 28, 2014

I think that the structure of Twitter (specifically: the fact that wheen you post, you post to an audience of all your pollowers) limits the range of possible types of 'conversatuon networks'. So don't take this post as all-inclusive. Where the value lies, however, in the recognition that very distinct forms of networks can form in an environment where people have conversations, some dysfunctional (such as 'polarized crowds') and others less so (such as 'community clusters'). Via Brent Schlenker.

[Link] [Comment]

Moneyball
Jim Groom, bavatuesdays, March 28, 2014

People are beginning to notice the the proponents of learning analytics need to brink some new data and examples forward to support their case, as the old ones are not only, well, old, they have also been thoroughly discredited. Jim Groom comments, "a lot has happened since 2010. Mike Caulfield pointed out six months ago, and Michael Feldstein re-iterated, the research claims of the effectiveness of Course Signals to increase retention are deeply problematic."

[Link] [Comment]

The Game of Wrong, and Moral Psychology
John Holbo, Crooked Timber, March 28, 2014

Some interesting post-MOOC reflections on learning and moral psychology from Crooked Timber, another example I think of how a MOOC well done results in the creation of new knowledge, as opposed to the mere transmission of the old (this may not be so much true for the students of a Coursera course as it is for the instructor). Anyhow, the reflections cause in me some thoughts about the apparent contradition between two principles I have long held, the first of which is a form of utilitarianism, and the second of which is a version of Kant's principle that each person is inherently valuable. Silly problems like the Trolley Problem are designed not merely to test these principles but to drive a wedge between them. But in fact, the two principles are different aspects of the same principle of ethics, to my mind.

[Link] [Comment]