— In the News —
The success of the Internet of Things relies on the premise that all the data being generated by internet companies and devices scattered across the planet belongs to the organizations collecting it. What if it doesn’t? This is must-read interview with Alex “Sandy” Pentland from the MIT Media Lab about data ownership.
"The overeager adoption of big data is likely to result in catastrophes of analysis comparable to a national epidemic of collapsing bridges." That sentiment might be easy to dismiss if it wasn't coming from someone as highly regarded as Michael I. Jordan, who is one of the world’s most respected authorities on machine learning. This is a thought-provoking read.
— Tools and Techniques —
At some fundamental level, no one really understands machine learning. Humans do really well with two and three dimensions, not thousands. But by directly observing what is actually happening, step by step, this tutorial provides an opportunity to understand neural networks in a much deeper and more direct way.
— Resources —
Last week, experts from across the data world came together in New York City for Strata + Hadoop World. Here's a collection of notable keynotes, interviews, and insights from the event. For a larger selection of presentations and interviews, see the Strata YouTube Channel.
The U.S. Chamber of Commerce Foundation recently hosted leaders in data innovation to discuss how data is changing the way businesses operate. Their report, The Future of Data-Driven Innovation, focuses on themes such as how data is driving the economy, the policies needed to foster data innovation, and how data can be used for good.
— Data Viz —
Linking Small Multiples can be a very effective way of visually exploring a large dataset. Here’s a great tutorial about using this technique, along with demos, sample code, and lots of linked references. There are lots of ideas here. This is very well done.
This article is for all data enthusiasts. If you’re interested in sports data, you’ll obviously be interested. But if you’re not interested in sports data, you’ll want to check this article out anyway. The sports world has lots of data and lots of resources; making this a worthwhile domain to watch for fresh ideas.