A Data Science Central Community
Added by Capri on April 4, 2012 at 2:30pm — No Comments
While I was writing the last post I was wondering how long before my followers notice the mistakes I introduced in the experiments.
Let's start the treasure hunt!
1. Don't always trust your data: often they are not homogeneous.
A good data miner must always check his dataset! you should always ask to yourself whether the data have been produced in a…Continue
Added by Cristian Mesiano on April 4, 2012 at 2:10pm — No Comments
In the data mining process, where do data scientists like you and me add the most value? Is it in exploring the data, uncovering anomalies and seeing relationships between elements? In selecting transformations for the elements to improve their representations for modeling…Continue
Added by Daniel Graettinger on April 4, 2012 at 11:30am — No Comments
By Richard Smith, Director of SAMSI
The Statistical and Applied Mathematical Sciences Institute (SAMSI) announces a new call for proposals to organize a summer program or workshop focused on a topic within the broad spectrum of new or recent research themes covered by SAMSI, to take place during the summer of 2013. A…
Added by Vincent Granville on March 31, 2012 at 7:00pm — No Comments
The federal government is beginning a major research initiative in big data computing. The effort, which will be announced on Thursday, involves several government agencies and departments, and commitments for the programs total $200 million.…
Added by Vincent Granville on March 30, 2012 at 1:30pm — No Comments
Big data must really be big to get its own White House-sanctioned research and development push.
The White House Office of Science and Technology Policy will host a live…Continue
Added by Vincent Granville on March 30, 2012 at 9:30am — No Comments
I like to…Continue
Added by Vincent Granville on March 30, 2012 at 10:00am — No Comments
Added by Vincent Granville on March 27, 2012 at 10:30am — No Comments
When Justin Bassett interviewed for a new job, he expected the usual questions about experience and references. So he was astonished when the interviewer asked for something else: his Facebook username and password.
Bassett, a New York City statistician, had just finished answering a few character questions when the…Continue
It’s been quite a while since business intelligence was the sole province of high-level strategic planners and C-level executives. Today, BI tools and analytics insights are being used by increasing numbers of people to answer immediate tactical questions and identify optimal courses of action. This trend has been enabled and accelerated by the consumerization of IT and the “democratization” of tech knowledge. As…Continue
Hospitals laying off workers, blaming in on fewer patients and lower revenue... No statistical model was able to predict it, although the explanation is obvious.
After several years into Great Recession
Added by Capri on March 21, 2012 at 6:00pm — No Comments
Interesting stuff from the Analytics Magasine from INFORMS (The Operations Research Society):
Added by Vincent Granville on March 20, 2012 at 2:30pm — No Comments
Added by Vincent Granville on March 20, 2012 at 2:00pm — No Comments
First of all, our team is absolutely delighted that Judea Pearl won the ACM Turing Award, widely considered the "Nobel Prize in Computing". For well over a decade we've been exclusively committed to developing software based on Bayesian networks, a paradigm established…Continue
Added by Vincent Granville on March 19, 2012 at 8:00pm — No Comments
Guest post written by Shomit GhoseContinue
Added by Vincent Granville on March 14, 2012 at 8:19pm — No Comments
Added by Vincent Granville on March 14, 2012 at 8:00pm — No Comments
Interesting article published in IEEE Spectrum.
The queen of the sciences may someday lose its royal status
By ROBERT W. LUCKY / MARCH 2012…Continue
Added by Vincent Granville on March 10, 2012 at 3:33pm — No Comments
For statistical process control, a number of single charts that jointly monitor both process mean and variability recently have been developed. For quality control-related hypothesis testing, however, there has been little analogous development of joint mean-variance tests: only one two-sample statistic that is not computationally intensive has been designed specifically for the one-sided test of Ho: Mean2<=Mean1 and StDev2<=StDev1 vs. Ha: Mean2>Mean1 OR StDev2>StDev1 (see…Continue
Added by J.D. Opdyke on March 9, 2012 at 7:41am — No Comments
Three metrics can be used to segment the population of data scientists. Each metric has three levels: high, medium, low. Hence the 27 (= 3 * 3 * 3) types of scientists.
Here are the metrics in question: