Artificial Intelligence with Python

While stumbling across cyberspace looking for some interesting Python tools and tutorials I found this rather interesting webcast. The video is of Raymond Hettinger at PyCon 2009. Raymond describes the usefulness of Python with applying artificial intelligence and data mining. This talk is very interesting to see how useful of a tool Python can be to performing some relevant Operations Research tasks specifically with data manipulation and learning. As I have mentioned before Open Source software offers a lot in the way of Operations Research tools. I hope you enjoy it as much as I did.

Source:
http://blip.tv/file/1947373

Weekend fun: The perfect way to slice pizza

The New Scientist has a fun article about slicing perfect pizza, mathematician style. The basic premise of the article is to find out if there is an optimal way to slice the pizza so that the pie can be distributed among diners equally. Rick Mabry and Paul Deiermann of LSU have been trying to prove the hypothesis of the equitable slices. Of course it comes as no surprise as it all comes down to the cut. I'm not really sure what makes me more hungry, reading about proofs or about the pizza? Either way you slice it looks like fun to me.

Flaw of Averages followup: The Average Lie

The Simple Dollar has a great article about the The Average Lie. The reason why I bring up this article because a recent IEOR Tools blog commentator opined in an earlier post on the Flaw of Averages by Sam Savage. The commentator said Sam didn't bring up any good examples. Actually I think Sam did bring up a great example in personal finance of the fallacy of assuming x% average growth rate for the life of an investment vehicle. This article by The Simple Dollar is a great illustration of Sam's example.

After posting that article I could think of many different examples of the Flaw of Averages. Some that come to mind in industrial operations include average rate of production, average time to complete a task, average demand for a product. I have had to deal with each of these scenarios often having to come up with better suggestions to management to deal with the operational challenge other than using an average value. I would love to hear other great examples.

Social Networks and Operations Research

Based on a request from a blog reader Michael Trick has set up the OR-Exchange as a social networking way to get Operations Research related questions and answers. It is an experiment to see if it would be beneficial to the Operations Research community. For those of us that have used the common Usenet groups in the past we have recently seen it degrade into a chasm of spam. Hopefully with these new networking sites this will help remedy that networking need in the Operations Research community.

On another note Analytic Bridge and LinkedIn is also a very good networking and discussion social networking tool. There are many "Groups" that one can join at LinkedIn to join with common Operations Research interests. Here are links to the following social networking sites devoted to all things Operations Research.

LinkedIn INFORMS group

LinkedIn DataShaping Advanved Analytics group

Analytic Bridge

Flaw of Averages recap and INFORMS podcast

I talked about an article about Sam Savage and his new book the Flaw of Averages not too long ago. This past Friday INFORMS released a new podcast in their series the Science of Better interviewing Sam Savage. This is a fascinating interview about the theory in his book and the, unfortunate often, over simplification of decision analysis based on a simple average of a data population. I strongly suggest listening to this podcast as Sam has a unique way of presenting his theories.

Legal issues with Data Visualization: AT&T and Verizon "Map For That"

CmdrTaco of Slashdot has a recent entry on AT&T sues Verizon over "Map For That" Advertising. CmdrTaco has a good point in that this story is about a lawsuit over data visualization. Sure there has been litigation in the past with false advertising. Yet this lawsuit is pointing to a topic that Operations Research practitioners take a great interest.

Data visualization is often seen as a great tool for comparison. I remember first seeing this TV ad from Verizon and thinking how amazed the difference in coverage between the two carriers. Data visualization can have a great impact on the recipients of the information. Often times the contrasts can be very startling. In fact so startling to AT&T that they are willing to sue over the data visualization chart by Verizon.

If there is anything to learn from this is that Data Visualization might be more important than we take for granted in Operations Research. As practitioners we often are trying to determine ways to improve our soft skills to help implement our ideas to management. Data Visualization can be a great tool in our tool bag to help deliver important results to the decision makers. Perhaps software applications could be developed to transfer output results of an optimization program to a data visualization chart. I am starting to get more interested in Data Visualization the more I think about it.

GLPK 4.40 released today

Andrew Makhorin, lead developer of the GNU Linear Programming Kit project, released the latest version of GLPK 4.40 today. Some of the following enhancements include...

• New API routines. Mostly with graphs.
• New branching and MIP heuristics
• And of course some bug fixes

As described by Andrew, GLPK is intended for solving large-scale linear programming, mixed integer linear programming, and related problems. GLPK is written in ANSI C and is organized as a callable library.

IEOR Tools also had a feature on GLPK on how to use it in the Windows environment as IEOR Tools tutorial: Using GLPK in Windows.