KDnuggets™ News 10:n11, May 19

Features

Poll Results: Free, Open Source Tools Grow in Popularity - May 17, 2010.
RapidMiner, other open source/free tools gain in 11th Annual KDnuggets Data Mining / Analytics Tool poll
New Poll: Largest Data Analyzed? - May 18, 2010.
What was the largest database or dataset you data-mined / analyzed? Please vote
KDD-2010 Call for Participation, July 25-28, Washington, DC - May 17, 2010.
KDD-2010 is the premier international forum for data mining researchers and practitioners. KDD-2010 will feature keynote presentations, oral paper presentations, poster sessions, workshops, tutorials, panels, exhibits, demonstrations, and the KDD Cup competition.
Most viewed items for week May 9-15 - May 16, 2010.
News: New Poll: Data Mining/Analytic Tools Used; Why F# is the language for data mining
Top Jobs: Statistical Analyst with Web 2.0 Focus at Fitzgerald Analytics and Bundle Corporation; Netflix - Director of BI and Analytics
Most viewed items for week May 2-8 - May 9, 2010.
News: New Poll: Data Mining/Analytic Tools Used; Facebook F8: One graph to rule them all
Top Jobs: Director of Data Analytics at SpeedDate.com; Predictive Modeler at Allstate Insurance Company
Competition: What caused May 6 stock market collapse? - May 18, 2010.
AnalyticBridge competition: investigate the spectacular stock market collapse of May 6, 2010

Courses (see also All Courses)

SUMMER SCHOOL NN2010 - May 11, 2010.
the school will explain relevant NN paradigms, including multilayer perceptrons (MLP), radial basis function networks (RBF) and support vector machines (SVM) used for classification and regression tasks, illustrated with applications to real data

Software (see also All Software)

Fast classifier design and deployment with PRSD Studio 2.1 - May 18, 2010.
PRSD Studio empowers industrial and academic researchers to quickly design robust statistical classifiers in Matlab and rapidly deploy their solutions in custom applications using a DLL library.
Salford NEW Integrated Suite, Training, Celebration - May 12, 2010.
SPM (Salford Predictive Miner) includes CART(r), MARS(r), TreeNet(r), and RandomForests(r), along with powerful new automation and modeling capabilities not found elsewhere. Celebrate in San Diego on Jun 23!
Rapid-I launches extensions for RapidMiner 5.0 and RapidAnalytics - May 12, 2010.
Extensions optimise application integration, connection with the web and the integration of user-generated content
Revolution Computing relaunches as Revolution Analytics - May 7, 2010.
Company launches with free enterprise software for the academic community and roll out of a new R community website: inside-R.org

Jobs (see also All Jobs)

Data Mining Analyst at Waterfront International, Toronto, Canada - May 12, 2010.
Waterfront International is a Toronto-based financial consulting firm, specializing in developing computer based statistical trading strategies.
Senior Director of Analytics at advertising agency, San Francisco, CA - May 11, 2010.
responsible for all analytic documents and output of the database management group
Forensic Analysis Mgr, SIU at Travelers Insurance, Hartford, CT - May 11, 2010.
Responsible for the management and development of a highly technical and analytical SIU staff in the development of new capabilities and finding business solutions through the use of technology and analytical techniques
Statistical Analyst with Web 2.0 Focus at Fitzgerald Analytics and Bundle Corporation, New York, NY - May 11, 2010.
Fitzgerald Analytics, a consulting firm focused on strategic analytics, and Bundle Corporation, a fast-growing Web. 2.0 venture focused on developing peer based financial data sets, are jointly seeking a database modeling and statistical analyst
Netflix - Director of BI and Analytics - Product Development - 3487 at Netflix, Los Gatos, CA - May 11, 2010.
We are searching for a business intelligence (BI) and analytics veteran to support the Netflix Product Development Team.
Quantitative Researcher/Data Miner at iProspect, Mumbai, India - May 10, 2010.
researching, analyzing, and reporting on data generated through iProspect's SEO and SEM services offerings, and client provided data; conference calls with clients required
Quantitative Researcher/Data Miner at iProspect, Chicago, IL - May 10, 2010.
responsible for researching, analyzing, and reporting on data generated through iProspect's SEO and SEM services offerings, and client provided data.
Senior Analytics Engineer at Turiya Media, San Francisco, CA - May 5, 2010.
You will drive all facets of the data mining and predictive analysis, including the collection, normalization, and subsequent analysis of the large volume of data required for this process

Academic/Research positions

Research Scientist in Educational Data / Text Mining & Statistical Learning at Nexus Research and Policy Center, San Francisco, CA - May 12, 2010.
work at the newly established Nexus Center; help execute Nexus's annual research agenda examining the processes and effects of teaching and learning at both proprietary and traditional institutions of higher education.
New MSc in Analytical Business Intelligence at BME, Budapest - May 18, 2010.
A new MSc course about Data Mining and Knowledge Discovery starts next semester at the Budapest University of Technology and Economics (BME)

Publications

Trust your gut? Or Data? - May 18, 2010.
Here are a couple of very interesting articles that highlight making decisions using your "gut" vs relying on data.
Mining and Analyzing Online Social Graph Data - May 17, 2010.
an interesting talk by Drew Conway, PhD student at New York University, Dept. of Politics, an expert on social networks)
What about Analytics in Social Media monitoring? - May 16, 2010.
Listening posts monitor the "chatter" that is occurring on the Internet in blogs, message boards, tweets, etc
Data Mining Data.Gov: Statistics from FEC Candidate Summary Files - May 13, 2010.
Since Primary Election season is ramping up in the US we'll demonstrate Solr's Stats functionality using some data from from Data.Gov
Data Inc. profiles data-driven companies - May 12, 2010.
The hit forecaster uPlaya analyzes songs against an ever evolving databank of past and present musical hits to estimate a song's potential for commercial success.
The Importance of Straight Data - May 11, 2010.
I illustrate the topic sentence by giving details of what to do when an observed relationship between two variables depicted in a scatterplot is masking an acute underlying relationship
Microsoft Attempts to Predict the Future - May 11, 2010.
One experimental program, called Predestination, collects and processes data about urban traffic patterns, the behavior of drivers, etc and recommends best route to where you are likely to go
An Interview with Revolution Analytics CEO Norman Nie - Part 2 - May 10, 2010.
even with 2M users of R, the challenges for commercial R purveyor Revolution Analytics (formerly REvolution Computing) are significant.
Why F# is the language for data mining - May 7, 2010.
F# is Succinct, static typing at the same time.

News Briefs

New book: "Bursts" - Can human behavior be predicted? - May 5, 2010.
An interesting experiment is associated with the launch of the book: the whole text is available on the book site, just as it will appear in print. But each word is covered by a rectangle.
IBM's New Analytics Software Interprets Slang and Emoticons - May 14, 2010.
Here's an example of a sentence that would be tough for most software to interpret: "The iPad's great...if you don't care about Wi-Fi. :-p"
IBM Debuts New Social Media Analytics Tool - May 13, 2010.
IBM is introducing a new social media monitoring tool, which will measure consumer sentiment from data gathered on Twitter, blogs and other web services and networks.
Researchers study behavior in social networks - May 12, 2010.
interesting observations from Cameron Marlow, manager of the Facebook data science team and D.J. Patil, chief scientist at LinkedIn.
UT Dallas behavioral data analysis system - May 12, 2010.
the system focuses on identifying potential Internet threats, but it has broad applications
Jiawei Han Honored by IEEE Computer Society with 2009 W. Wallace McDowell Award - May 8, 2010.
He received the award "for significant contributions to knowledge discovery and data mining."
Nielsen Acquires Online Video Analytics Company GlanceGuide - May 7, 2010.
Analytics and audience measurement giant Nielsen has acquired video analytics company GlanceGuide
N.Y. bomb plot highlights limitations of data mining - May 5, 2010.
Like weather forecasting, data mining can predict major storms but not where each raindrop will fall
Google invests in firm that tries to predict the future - May 5, 2010.
Google has invested in a startup company that claims to be able to predict the future.

CFP - Calls for Papers (see also All CFP)

Discovery Science 2010, due May 19
SIGSPATIAL GIS 2010: Advances in Geographic Information Systems, due Jun 24
Computational Intelligence in Knowledge Discovery , due Jun 26
ICDM '10, due Jul 2
ICDM'10 Workshop on Domain Driven Data Mining, due Jul 23
ICDM'10 Workshop on Visual Analytics and Knowledge Discovery, due Jul 23
User Interfaces for Recommender Systems, due Nov 5

Quote

Here's a simple, cheap experiment: the next few times you make a snap decision or judgment where your gut is your best friend, take 20 seconds to send yourself an email or text briefly describing what you did and why. Quit after doing it maybe 20 times. Then look at those messages from the vantage of a week later. I promise you'll be surprised.

Michael Schrage, Tell Your Gut to Please Shut Up, http://blogs.hbr.org/schrage/2010/05/your-gut-is-overrated-really.html