— In the News —
If you're interested in machine learning, you'll definitely want to check this out. At over 360 upvotes, this is one of the most popular Machine Learning posts on Reddit in quite awhile. There are lots of insights here regarding industry trends, career paths, and implementations.
How much do people around the world think their data is worth? If your business depends on access to user data, this is an important question since consumers are becoming increasingly savvy about how their data is used. This in-depth article dives into ways to provide value for users' data while building trust along the way.
I'm not sure that detecting "human-like" intelligence really gets at the heart of the matter but that is what Turing was going for. The Turing Test has been getting a lot of well-meaning attention lately and this is an interesting read that includes lots of linked references.
— Tools and Techniques —
There were a couple of great conferences this past week that are worth noting. OpenVis is definitely one of the best data visualization conferences anywhere and all of the presentations from the 2015 OpenVis conference have just been put online. If you're interested in data visualization, these are MUST SEE presentations. Then there was PyCon in Montreal which was an entire week of Python talks and tutorials. This 3 hour tutorial by Jake VanderPlas called Machine Learning with scikit-learn is particularly well done.
The final installment of a three part series that explores the massive landscape of tools that are available to data scientists. This week's focus: Data Applications - where the "sexy stuff" like predictive analysis, data mining and machine learning happen. This is the part where you take all your data and do something really amazing with it...
— Resources —
Here's a growing resource that's worth paying attention to. Stamen Design is collecting the best freely-available terrain datasets for the entire world and developing cloud-based tools to process and work with that data. Check out their project on Github at the link above and definitely check out their Tumblr too. With funding by the Knight Foundation, this promises to be a very interesting resource.
Nice collection of public datasets on GitHub. There are currently 271 datasets listed, covering a broad range of domains. This collection is well organized, has been starred 2300 times and it looks like it's regularly updated.
— Inspiration —
So cool! This is an awesome new episode of NOVA that describes itself as “a mathematical mystery tour - a provocative exploration of math's astonishing power across the centuries.” Highly recommended.
The recently released Chef Watson cookbook is "a revolutionary display of the creative collaboration of man and machine.” Over the past three years the Institute of Culinary Education (ICE) has been collaborating with IBM researchers, testing the limits of ingredient pairings and developing strategies to implement cognitive cooking in the culinary arts. The short videos on this ICE blog really round this out into a fantastic article about the project.
— Archive Pick —
Great article about building a music recommendation system using convolutional neural nets. This goes deep if you want details but there's also a high-level path through the article that provides a great overview of how it works. Highly recommended.
— Career —
In addition to overall rankings, this article by CareerCast has lots of info including pay ranges, hiring outlook, and insights about work environments and typical stress levels. You're in luck if you like working with data - there are several data-related jobs in the top 10. Check out the report to make sure you're on the right path.
Great insights here for people interested in making the move from academia to industry. If you’re in that camp, this is a MUST READ.