KDnuggets™ News 10:n21, Sep 9

Features

Becoming a data scientist - Sep 8, 2010.
Alex Kamil gives eight points from his perspective as an undergraduate student. Many of these reference statistics and math, and Kamil provides an excellent list of papers, websites and technologies to tinker with.
EmotionML: Will computers tap into your feelings? - Sep 6, 2010.
W3C works on EmotionML, formalizing emotional states in a computer-readable way. Using emotional states, called "affective computing", can make computer interactions very different. Avatar faces could show their human master's expression during computer chats.
PAW DC Early-Bird Ends Friday; PAW London Agenda Announced - Sep 8, 2010.
Predictive Analytics World DC, Oct 19-20, 2010, includes speakers from Canadian Imperial Bank of Commerce, Life Line Screening, Merkle, MetLife Auto & Home, Miles Kimball, Monster, Paychex, Xerox PARC. PAW London, November 15-16, 2010, delivers a comparable line-up.
M2010 Data Mining Conference - get KDnuggets Discount - Sep 2, 2010.
Attend M2010 and learn latest techniques to give your company a competitive edge. Through keynote presentations, case studies and technical presentations, you'll see real-world examples of how others are using analytics to get a competitive edge.
SAS says it is not WPS victory - Aug 27, 2010.
Court finds WPS Infringed on SAS Copyrights; refers case to European Court of Justice
Most viewed news, jobs in August - Sep 5, 2010.
P != NP Proved by HP Labs researcher?; The Most Comprehensive Social Media Analytics Platform List;
Top jobs: Research Scientist, Data Mining & Machine Learning at Siemens Corporate Research; Researcher, Predictive Analytics at Xerox Research Center, Webster, NY;
Most viewed items for Aug 29 - Sep 4 - Sep 5, 2010.
Social media analytics - 2 new tools; 5 indispensable IT skills of the future;
Top jobs: Research Engineer - Targeting and Optimization at Amazon.com, Seattle, WA; Research Scientist, Forecasting Team at Amazon.com, Seattle, WA;
Additions to KDnuggets Directory since Aug 1 - Sep 1, 2010.
new in companies, datasets, education, meetings, software, solutions, websites
Most viewed items for Aug 22-28 - Aug 30, 2010.
Poll Results: half of SAS users consider switching; P != NP Proved?;
Top jobs: Senior Manager or Manager - Statistician at Global consulting organization, 8 locations; Statistician, Ford Credit;

Webcasts (see also All Webcasts)

Data Mining: Failure to Launch [ Sep 9 Free Webinar ] - Sep 8, 2010.
Attend this free webinar to learn how to get started with data mining and overcome both strategic and tactical limitations that cause data mining projects to fall short of their potential.

Software (see also All Software)

WekaSharp: An F# wrapper for Weka - Sep 7, 2010.
This post is for end users, who want to use this wrapper to perform data mining tasks in .Net.
Sizing samples - how much data is enough? - Sep 6, 2010.
MIT researchers have shown that graphs shaped like stars and chains establish, respectively, the worst- and best-case scenarios for computers doing pattern recognition.
SAS launches do-it-yourself predictive analytics for business users - Sep 3, 2010.
SAS Rapid Predictive Modeler automatically guides users through a behind-the-scenes workflow of data preparation and data mining tasks, enabling them to generate their own models and derive insights. It runs from either SAS Enterprise Guide or Microsoft Excel Add-in.
Social media analytics - 2 new tools - Aug 31, 2010.
As new tools develop that are free to use, it becomes harder for paid tools like Radian6 to justify their charges. There's just so many free tools out there that do a good job and make social analytics accessible for any size business.
WordStat 6.1: new multilingual user interface, better linguistic support, faster - Aug 30, 2010.
WordStat is a flexible content analysis and text mining tool that quickly extracts themes, trends and patterns from large collections of text. It can analyze documents in most human languages.
New Google Service Will Decide For You - Aug 28, 2010.
Google's service is particularly good at distinguishing between different languages
Start-up offers free social media analytics service - Aug 26, 2010.
Viralheat offers Web users and publishers a limited slice of the data it collects from the social Web for paying customers

Jobs (see also All Jobs)

Analytical Modeling Staff Scientist-10002174 at SAS, San Diego, CA - Sep 8, 2010.
As a member of the Financial Services Modeling team, you will analyze customer data and build high-end analytical models for solving high-value business problems, such as credit card fraud, credit risk, network security, tax fraud detection, and revenue and collections optimization.
Analyst/Associate, Business Intelligence at J.P. Morgan Asset Management, New York, NY - Sep 8, 2010.
to work within the BI team to develop a strategic framework across all of our sales channels and within distribution strategy.
Advanced Analytics Math Modeler-1002995 at Dow, Michigan, MI, USA - Sep 8, 2010.
applying math for decision support - assist businesses and functions. There are two basic frameworks to support - data mining and then mathematical modeling. Data mining is generally used, from a Six Sigma perspective, in 3i and Opportunity Identification projects.
Machine Learning Engineer at eBay Inc., San Jose, CA - Sep 3, 2010.
apply your software dev. and machine learning skill to real world problems that require defining and implementing state of the art techniques for mining and understanding massive amount of unstructured (free form text description) , semi-structured (user behavior) and structured (catalog) data.
Senior Analysts at Argus, San Francisco, CA - Sep 3, 2010.
a B2B company providing strategic and analytic solutions to major financial services companies (Credit Card issuers) in the US, UK and Canada.
Research Scientist at OpenX, Pasadena, CA, USA - Sep 1, 2010.
OpenX's Ad Server and Market analytic solutions. The research team is responsible for investigating, designing, prototyping, and reviewing analytic solutions in support of OpenX's Ad Server, Market, and other offerings. The research scientist will be responsible for one or more analytical assignments within a supportive environment.
Research Scientist, Forecasting Team at Amazon.com, Seattle, WA - Sep 1, 2010.
we develop sophisticated algorithms based on the latest statistics tools and data mining methodologies. You will have an opportunity to work on an incredibly complex mathematical problem, with a large element of unpredictability.
Predictive Modeler at Allstate Insurance Company, Menlo Park, CA and Northbrook, IL - Aug 31, 2010.
apply economic, quantitative, statistical, and actuarial skills in the development of new pricing and financial models which contribute to Allstate's growth and profitability.
Trade Marketing Director/Retail Analytics Expert at ProPhase, Doylestown, PA - Aug 31, 2010.
form insightful marketing recommendations from analyzing IRI, scanner and quantitative retail data. You see the story in the numbers. Determining the ROI and cost of doing business per channel is your strength.
Research Engineer - Targeting and Optimization at Amazon.com, Seattle, WA - Aug 30, 2010.
outstanding Sr. Research Engineer to join the Targeting and Optimization team within Amazon.com's Display Advertising business
Distributed Platform Engineer at Benefitfocus, Charleston, SC - Aug 25, 2010.
original algorithm design, an understanding of Unix and performance issues, and the ability to work with our research and operations groups. The candidate must be able to contribute to the overall architecture of our product and take individual ownership of large components of the system
Semantic Intelligence Engineer at Benefitfocus, Charleston, SC - Aug 25, 2010.
focus on writing software for aggregating health benefits data from distributed sources via raw, linked data or other techniques; managing one or more stores of aggregated RDF data for distributed indexing, analysis, and visualization; and creating frameworks that will operate over massive distributed RDF stores
Data Scientist at Benefitfocus, Charleston, SC - Aug 25, 2010.
perform data analysis, develop algorithms and predictive models, and present technical information to a broad audience. You will apply algorithms and modeling tools to a variety of data, and develop critical solutions for multiple areas of business

Academic/Research positions

2 Postdoc/Assistant Professor positions in NLP at Sapienza U. of Rome, Rome, Italy - Sep 8, 2010.
participate in a frontier research project on multilingual semantic processing and will work in a stimulating environment of a leading and highly-active research team including 2 faculty members, 2 research associates and 3 Ph.D. students.

Meetings (see also All Meetings)

SmartContent: Content Analytics Conference, Oct 19, New York City - Sep 2, 2010.
We have a stellar array of speakers lined up for Smart Content, a new conference, that tackles digital transformation: analytics solutions that enrich and monetize Web, enterprise, and social content.

Audio/Podcasts/Video

Tutorial by Christian Thurau about "Data Mining in Games" [Video] - Aug 31, 2010.
Modern games are generating massive amounts of data that can help developers for tasks such as user modeling, behavior analysis, matchmaking, or playability testing. This tutorial will give some background on modern data mining techniques and explain its application to games.

Publications

Is the AUC the Best Measure? - Sep 8, 2010.
The area under the curve (AUC) that relates the hit rate to the false alarm rate ia a way to measure predictive modeling accuracy. This paper examines the problems with using AUC.
Economist: Untangling the social web - Sep 7, 2010.
From retailing to counterterrorism, the ability to analyse social connections is proving increasingly useful
Cyber Security Data for Experimentation - Sep 3, 2010.
NSF wants to fund a project to combine a data-mining expert with the Tor data and see what insights can be extracted from it. If you are that data-mining expert, please contact us.
Think the Answer's Clear? Look Again - Sep 1, 2010.
Scientist at Work: Dr. Donald A. Redelmeier. He can totally against intuition, and come up with a beautiful finding, such as using cell phones while driving increases collision risk 4-fold, and benefits from changing lanes are illusory.
Location Determines Social Network Influence - Sep 1, 2010.
Number of connections less important than proximity to core, according to CCNY team
To Data Mine or Not to Data Mine in the Fight Against Terrorism - Aug 27, 2010.
The 2-day conference entitled Data Mining and Human Rights in the Fight Against Terrorism was held June 10-11, 2010 in Zurich. The organizers had previously conducted a study on data mining in the context of counter-terrorism, and this became the basis for the conference.
Analytics skills in demand - and analytics pros demanding top salaries - Aug 25, 2010.
As more and more companies deploy predictive analytics tools and other data analytics software and begin filling the ranks of their analytics team, the pool of available talent is shrinking and hiring costs are growing

News Briefs

InferX Awarded a Contract to Perform 'IED Data Analyses Using Predictive Analytics and Advanced Visualization' - Sep 7, 2010.
Company Shows How Next-Generation Predictive Analytics Products and Solutions Help Address Critical National Security Issues Related to IEDs.
comScore Acquires Nedstat; Alterian Acquires Intrepid - Sep 6, 2010.
comScore acquired Nedstat, Amsterdam-based provider of web analytics and innovative video measurement solutions. Alterian acquired Intrepid, an international market research and social media analytics consultancy.
UT, ORNL collaborate to improve accuracy of climate change models - Sep 6, 2010.
While some climate variables, such as temperatures over atmospheres and oceans, are relatively easy to predict with models, more extreme variables, such as hurricanes, are not, but they may be more important for climate change impacts and policy.
Real-Time Analytics Provider Chartbeat Raises $3M from All-Star Investors - Sep 6, 2010.
Series A is led by Index Ventures and includes Ron Conway, Jeff Clavier, Jason Calacanis, Chris Sacca's Lowercase Capital and Betaworks.
Mind-Boggling: Deloitte Stumped By Low Awareness On Business Analytics - Sep 1, 2010.
In a recent Deloitte poll, 33% of 1,900 technology executives and business professionals, either didn't know if their organization utilized business analytics - or even if they had business analytics capabilities at all.
The Unica Story: Yuchun Lee's Journey from MIT Blackjack Team to IBM Acquisition - Sep 1, 2010.
Because of his past, Yuchun Lee is legally barred from certain establishments. Luckily, IBM is not one of them.
5 indispensable IT skills of the future - Aug 31, 2010.
Demand will be high for the ability to analyze BIG data, and also work with business units to define what data is needed and where to get it. #2 Skill: Understanding Risk.
Stopping crime before it starts - Aug 27, 2010.
Sophisticated analysis of data can sometimes tell police where criminals are headed. LAPD plans to use such tools.
Visual IQ Announces $3 Million Investment by Fog City Capital - Aug 26, 2010.
Visual IQ provides software and services that help its clients turn data-driven insights into actions that deliver better returns on their marketing dollars,
SAS(r) Text Analytics named Trend Setting Product for 2010 - Aug 26, 2010.
SAS recognized for innovation and identifiable technology breakthrough
Microsoft Top Three hottest new majors for a career in technology - Aug 26, 2010.
They are Data Mining/ML/AI/NLP, Business Intelligence, Web Analytics/Statistics

CFP - Calls for Papers (see also All CFP)

PAKDD 2011, due Sep 19
Machine Learning in Online ADvertising, due Oct 23
AI, Wikipedia and Semi-Structured Resources", due Oct 31
Advances in Social Network Analysis and Mining workshops, due Nov 1
Social Computing, Behavioral Modeling, & Prediction , due Nov 6
Machine Learning and Data Mining, due Mar 1
Mass Data Analysis of Images and Signals in Medicine et al, due Mar 1

Quote

Discovery consists of seeing what everybody has seen and thinking what nobody has thought.

Albert Szent-Gyorgyi (1893 - 1986)