Screenshot

I read this story wanting to understand if the data mining they’re doing is really appropriate for making individualized statements in the way they are claiming when they suggest that hospitals will get risk assessments based on patient shopping data through credit cards, store cards, etc. Will receiving doctors get sufficient training in the ways in which these predictions are like and unlike the predictions that medical tests make about health risks?

Additionally, I read through the list of hypothesized triggers for heath risks and they seem to bank on the idea that everything I’m purchasing is for myself. Just in the first paragraph the article suggets issues if “ou’ve let your gym membership lapse, made a habit of picking up candy bars at the check-out counter or begin shopping at plus-sized stores” which could nicely match my patterns this year of regularly picking up candy for the lab and buying some clothes for an elderly relative who can’t get out to stores as easily anymore. The majority of the time I buy donuts, I do not actually eat any of the donuts – and I have colleagues whose donut-eating patterns may more closely match my shopping trends than theirs.

And then you get this statement: “While the hospital can share a patient’s risk assessment with their doctor, they aren’t allowed to disclose details of the data, such as specific transactions by an individual, under the hospital’s contract with its data provider.”

I’m not sure if that’s good — hooray for not sharing personal details! — or worse — so the computer says I’m at risk but we can’t sit down and talk about whether the patterns it’s identifying are real. And, going back to my opening question – what sorts of algorithms are being used and given that, what sorts of conclusions are even valid to draw.

With the semester over, I’m looking to what projects I’ll be taking on for the next couple of months, and I know many of my students are as well. Here are a few fun options people may want to consider, particularly focused on opportunities to get involved with data analysis:

• Try out Mortar, an open source recommendation engine platform, described a bit here
• Take part in Kaggle’s Acquire Valued Shoppers Challenge, playing around with their 22GB of anonymized sample data of 300,000 shopper’s data; this is a massive dataset, Revolution Analytics offers some pointers to tools to help you manage it (you can also check out Kaggle’s other predictive modeling competitions for problems in other domains)
• Get hooked on playing 2048 and then use this R framework to build your own 2048 solver that you can watch try out different strategies
• Check out this Udacity course on Exploratory Data Analysis using R from four Facebookers

I hadn’t run into the unsolved Dorabella cipher before (that I remember). If you enjoy such things I highly recommend this account of it, with its many proposed decryptions that make clear why one of the conditions for a verified solution is that it “be self-evident”. It’s an excellent example of why decyphering without context is hard (maybe impossible?). And I enjoy the proposed solution that takes encryptions errors into account as a possibility, considering that it was done by hand, and by someone considered prone to such errors.

Nothing like avoiding end-of-the-year physical cleanup with end-of-the-year virtual cleanup! I finally got around to reading this detailed description of how Bitcoin works, recommended by Schneier on his weblog, and I need to hang on to this for next time I’m teaching security. From a teaching perspective, it does a nice job of showing how all of the various types of cryptography come together in an interesting way in this protocol. This is the part that always seems sort of wild to me:

The idea is to make it so everyone (collectively) is the bank. In particular, we’ll assume that everyone using Infocoin keeps a complete record of which infocoins belong to which person. You can think of this as a shared public ledger showing all Infocoin transactions. We’ll call this ledger the block chain, since that’s what the complete record will be called in Bitcoin, once we get to it.

The article does assume you have some cryptographic background, but I suspect that reading along as far as you can through the article would at least explain what some of the problems that Bitcoin has to solve are. A nice read with a cup of tea on a snowy day, especially if you’re a student getting your brain back into gear for school in another day or two!