A first collection of active data science resources. This is a short list of blogs that are still active, there a larger number of inactive data science blogs, that are still very valuable if anyone has the time to go back through all the previous posts.
Applied data science Blogs:
Obsession with regression,
http://obsessionwithregression.blogspot.com/ a data science blog run by a PhD candidate at Stanford (she'd be a great person to interview for anyone interested in data science, extremely high performer. I know her from high school I'm sure she'd be willing to help anyone....as long as she remembers to return your message). This blog has a lot of her side projects, mostly on analysing twitter data and text files. In the process she covers a lot of useful data visualization tools.
Kaggle Projects:
http://benanne.github.io/2015/03/17/plankton.html , these posts cover in detail the process a group of grad students used to apply a set of deep learning algorithms to two kaggle competition (one of which they took 1st place). A very good resource to see first-hand what goes into a successful data analysis.
Research / mathematically intensive data science blogs:
Geomblog:
http://blog.geomblog.org/ A blog by a professor, covers a varied set of topics at a research level.
I'm a bandit:
https://blogs.princeton.edu/imabandit/ A blog by Sebastian Bubeck, a brilliant research who just left Princeton to join Microsoft research. Covers a large number of topics, many in convex geometry and optimization, but has numerous valuable resources about multi-armed bandits & statistical learning theory.
The Spectator:
http://blog.shakirm.com/ This is a new blog by one of the cofounders of google-deepmind. It covers a set of mathematical tricks to improve machine learning algorithms.
Short, fat matrices:
https://dustingmixon.wordpress.com/ A professor at Air Force Institute of Technology. Covers his research, it's covers topics that are very important in the design of new machine learning algorithms and understanding statistical learning, but has little relevance for applied data analysis.
Nuit Blanche:
http://nuit-blanche.blogspot.com/ A blog with daily posts collecting new software implementations of algorithms, links to videos & tutorials, and new papers in the field.