Follow the Money

In a beautiful video, researchers explained how they were able to track dollar bills and used that information to calculate borders of social networks. Some of the borders seem to follow state lines, others follow geographic landscapes (e.g. the Mississippi). What was especially nice about this video are the animations which make it easy to understand what calculations are being done.

The equation for a city

I found this article in the New York Times very fascinating about two physicists who decided that they wanted to create an urban science which rethinks the way that we look at urban planning. They found that there are simple relations between certain parameters of a city. With a few equations they can make predictions about all cities that are remarkably accurate. I'll let the article speak for itself, but the following quote seems to summarize the

“I can take these laws and make precise predictions about the number of violent crimes and the surface area of roads in a city in Japan with 200,000 people. I don’t know anything about this city or even where it is or its history, but I can tell you all about it. And the reason I can do that is because every city is really the same.” After a pause, as if reflecting on his hyperbole, West adds: “Look, we all know that every city is unique. That’s all we talk about when we talk about cities, those things that make New York different from L.A., or Tokyo different from Albuquerque. But focusing on those differences misses the point. Sure, there are differences, but different from what? We’ve found the what.”

Math knows who and what you love

If you go to the website OKCupid you will see that it self identifies as the 'best dating site on Earth' and this is probably an indication of what online dating is like: a little self promotion never hurt anyone. Right?

Another tagline that they use: 'we do the math to get you dates'

The site matches its users by the use of online quizzes which makes it different than the online dating sites that I've seen. The data collected by this site is probably quite interesting to read. A blogger for the site by the name of Christian Rudder is identified as a guitarist and a math guy in an interview on NPR. If you look at his Wikipedia page he is labeled as American musician, humorist, and entrepreneur.

This story was about the analysis that Christian Rudder (and his team) did on the data available through OKCupid. Admittedly it was just for fun, but I bet this data would be of interest to marketers. They first tried to identify the likes of groups that one would normally try to stereotype. They didn't find anything interesting because these lists looked very similar independent of what group one was looking at. But then they tried something interesting and learned a good mathematical definition of what a stereotype is: a 'like' that a group has that is not common to the whole population.

What did they find? From the NPR article:

For example, top interests for white men included Tom Clancy, Van Halen, golfing and Harley Davidson. Top interests for black women were soul food, The Color Purple, gospel and Alicia Keys.

I know that this sounds kind of sad, but really it is identifying what we call a stereotype and why.

The most recent posting to the OKCupid blog was about statistics about gay and straight sterotypes and other statistics from data which seems to involve a lot of really good sociology research.

Running and Food, calculation of calories

A graduate student in the MD/PhD program of Harvard/MIT division of health, Benjamin Rapoport, realized that the phenomenon of 'hitting the wall' while running a marathon was an effect of a body not storing enough carbs. He decided to create a mathematical model to try to determine a formula for the carbohydrates that someone should store before running a marathon. There have been several reports in the news lately of this very useful application of mathematics. It seems as though Benjamin was inspired by his love of running marathons (he has run 18) and his background in the health sciences and mathematics.

His model allowed a group to build an 'endurance calculator' to determine safe running speeds over long distances.

NPR,
New Scientist
Outlook Series.

Why didn't the robot cross the road? Because he couldn't see.

In the economist's Babbage podcast from October 23 (their first), there was a discussion on computer vision. They mention that the reason we don't see more robots in our everyday lives is that they really aren't able to see where they are going. While they can be trained to interact with specific types of objects, developing an algorithm for computer vision has its own challenges. They mention that the most successful computer vision algorithms out there begin with edge detection and try to build up to larger objects.

There is an enormous community working on the problem of computer vision, so it seems like the Economist's coverage cannot give a very complete overview of the state of research in this area. What I think they can do is talk about some of the most complicated tasks that computer vision can and cannot accomplish.

Shake, rattle and roll

I heard this beautiful example of an application of mathematics. How fast does a dog need to shake in order to dry off? This is an excellent example where it is both surprising that you can apply mathematics to solve that problem and it is also surprising that the answer ends up being relatively simple that you can summarize it in a short story.

This interview appeared on NPR but there were several news organizations which carried the story which includes video: Discover magazine, Wired Science, Guardian, Physorg.com. You may also be interested in the paper that was posted under the arxiv and the corresponding video (a repository of math and physics papers).

Manufacturing is driven by technical innovation

In a story about manufacturing in the U.S., the Planet Money of NPR put together a remarkable story about why manufacturing in the U.S. succeeds. They mention that nearly 22% of all goods are manufactured in the U.S. This fact seemed quite counterintuitive to (my) conventional wisdom about the direction of the U.S. economy and the idea that manufacturing is moving or has all moved to China. While a lot of manufacturing is leaving the U.S., there is a reason why U.S. companies can compete and keep their manufacturing in the U.S.: innovation!

The story is presented by comparing two different factories, one that is failing because it produces something that can be made elsewhere for a lot less money, and a second factory that is succeeding because they produce a product that requires a large amount of innovation. This second factory makes electronic connectors and they need to come up with a new product nearly every year to stay ahead of the curve. There are engineers who work at this factory and talk to other engineers at other companies to try to solve their technical issues.

One line about this story made me sit up and take notice: "...if your parents can understand the product you make, there is a good chance that someone in China is making it a lot cheaper." The U.S. and Canadian economy is being driven by individuals who are able to come up with the new ideas to solve engineering and technical problems and invent new products. This requires mathematical and scientific skills that distinguish U.S. students. "This idea of the one guy in the back, having ideas, and then that fueling all the profits that allow a manufacturing business to succeed. That really is the model for U.S. manufacturing to succeed. U.S. manufactures that succeed make things, sure, but they really make their money by coming up with new ideas."