KDnuggets™ News 10:n27, Nov 23

Features (5) | Software (8) | Jobs (4) | Academic (5) | Meetings (3) | AudioVideo (1) | Publications (8) | NewsBriefs (11) | CFP (11) | Quote

Features

• New Poll: "Do not track" registry vs ad targeting - Nov 22, 2010.
  US Congress is currently discussing "Do not track registry", which will allow people to opt-out of tracking. Do you think such registry, if enacted, will significantly reduce ad targeting and behavioral advertising?
• Cloud analytics in test mode - Nov 22, 2010.
  cloud analytics is still in test mode, but its acceptance is rising - about as many people plan to use it in the next 6 months as have already used it.
• Most viewed items for Nov 14-20 - Nov 21, 2010.
  PredictiveDB (currently in alpha); Google Refine;
  Top jobs: Decision Scientist, Fraud Risk at Microsoft; Research Soft. Engr at UMD
• Predictive Analytics World March 2011, San Francisco - Super early bird rate ends Dec 7 - Nov 22, 2010.
  Predictive Analytics World - San Francisco - Super early bird rate ends Dec 7. Take advantage of the super early bird rate and make savings of $200 for Predictive Analytics World - San Francisco (March 14-15 at the Marriott Marquis).
• Analytics jobs exceed BI jobs - Nov 18, 2010.
  According to mega-jobs site indeed.com, analytics jobs have grown faster in the last 5 years than BI jobs

Software (see also All Software)

• Twitter Official Analytics in Alpha - Nov 22, 2010.
  Twitter Analytics (now in alpha) will let users see information like which tweets are most successful, which tweets caused people to unfollow, who the most influential users are that reply and retweet their messages.
• New Contest in TunedIT: Prediction of Biological Properties of Molecules. - Nov 17, 2010.
  The task is to predict an important biological property of molecules from their chemical structures.
• Software for statistically significant positive dependencies - Nov 17, 2010.
  Kingfisher program searches for the best non-redundant dependency rules, which express statistically significant positive dependencies in data.
• 11Ants Analytics Introduces Revolutionary Method of Building Ensembles - Nov 16, 2010.
  The new feature allows users to quickly build and evaluate thousands of ensemble permutations. Ensembles are recognized to increase the predictive power of models, and are a valuable tool in any data mining expert's arsenal.
• Google Cleans Up Messy Data with Refine - Nov 15, 2010.
  Google Refine is a power tool for working with messy data, cleaning it up, transforming it from one format into another, extending it with web services, and linking it to databases like Freebase.
• PredictiveDB is SQL-based predictive software (currently in alpha) - Nov 15, 2010.
  PredictiveDB is online software which uses SQL to learn, classify or predict from data. Currently in alpha
• Predixion Software Introduces Interface to Access and Run R Models in the Cloud - Nov 12, 2010.
  Predixion Insight(tm) - low-cost, self-service, cloud predictive analytics solution, offers plug-in for direct access to the R open-source statistical programming language
• Kaggle-in-Class launches with Stanford Stats 202 - Nov 11, 2010.
  Kaggle-in-Class is a new initiative that allows instructors to host data prediction and machine learning competitions for students.

Jobs (see also All Jobs)

• Mathematical Modeling and Optimization position at ExxonMobil, Corporate Strategic Research, Clinton, NJ - Nov 22, 2010.
  creative and resourceful individuals to join our research and development team to solve a multitude of challenging problems in the area of process modeling, experimental design or parameter estimation.
• Decision Scientist - Fraud Risk Analytics at Microsoft AdCenter, Bellevue, WA - Nov 18, 2010.
  analyze, optimize and implement computational rules used to monitor and control activity on the Commerce Transaction Platform (CTP). We are looking for a customer-focused, detail-oriented, results-driven analytic professional with great communication skills and a strong knowledge of databases and statistical techniques that also has a background in the payments and fraud prevention space.
• Research Software Engineer at U. of Maryland, Dept. of CS, College Park, MD - Nov 14, 2010.
  Implement an advanced machine-learning software package to support cutting-edge research; be responsible for all aspects of development: conceptualization, design, implementation, documentation, testing, debugging, external software integration and user-support.
• Predictive Modeler at DataQuick Information Systems, San Diego - Nov 11, 2010.
  DataQuick is a market-leading provider of real property information, often the first to bring new products to market. The Predictive Modeler works within the Analytics team building, deploying and maintaining predictive models and running statistical analysis.

Academic/Research positions

• Assistant/Associate/Full Professor of Business Analytics at U. of Tennessee, Knoxville, TN, USA - Nov 16, 2010.
  A primary research interest in developing analytic tools relevant to business is expected. Success publishing research in top journals is desired, along with an interest in collaboration with other College of Business colleagues and experience with finding solutions to business problems
• Information Systems Tenure-Track Junior Faculty Position at CMU, Tepper School of Business, Pittsburgh, PA - Nov 14, 2010.
  We have a particular interest in applicants with research and teaching interests in the development and application of machine learning techniques to business problems or in other related areas of business analytics.
• Tenure-Track Assistant Professor Position at Illinois Institute of Technology, Dept of CS, Chicago, IL - Nov 12, 2010.
  Applications are invited for a tenure-track assistant professor position in Computer Science beginning Fall 2011.
• Postdoc in Graph Mining and Network Analysis (GMNA) at Rutgers U., New Brunswick, NJ - Nov 10, 2010.
  for an outstanding individual to conduct research on graph mining, relational learning, and social network analysis. The project involves research on role formation and evolution in complex networks.
• Postdoctoral Researcher, Knowledgebase for Invertebrate Paleontology at U. of Kansas, Information and Telecommunication Technology Center (ITTC), Lawrence, KS - Nov 10, 2010.
  join a NSF-funded project developing a knowledgebase for Invertebrate Paleontology. Requires PhD in CS or related, with research experience in IR, machine learning, data mining, image processing/computer vision, or natural language processing.

Meetings (see also All Meetings)

• KDD-2011 Call for Papers and Participation - Nov 15, 2010.
  Abstracts due Feb 11 for 17th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD-2011), Aug 21-24, 2011, in San Diego, CA http://www.kdd.org/kdd2011/
• Nov 13 ACM Data Mining Camp Report - Nov 21, 2010.
  It was a big day for discussing hot topics in data mining, Mahout, parallel SVMs etc, and also a pretty big day for R.
• Registration now open for SAS Global Forum 2011 - Nov 9, 2010.
  April 4-7 at Caesars, Las Vegas featuring 2.5 days of learning and networking, SAS Global Forum 2011 is the premier event for SAS professionals worldwide.

Audio/Video

• Finding the right time to buy your plane ticket - Nov 17, 2010.
  Marketplace interviews Oren Etzioni, who developed data mining solution for predicting when to buy a plane ticket: The closer we get to the holidays, the higher airfares go. And we will pay them. But, timing isn't the only thing that we do wrong when it comes to picking a holiday flight.

Publications

• Accidental Statistician: Who Can Befitted of a Self-described Caption? - Nov 19, 2010.
  ... I once tried to read a book about it by someone called R. A. Fisher but I didn't understand it. He said "You've read the book so you better do it", so I said, "Yes sir".
• New Book: Networks, Crowds, and Markets, by David Easley and Jon Kleinberg - Nov 19, 2010.
  Drawing on ideas from economics, sociology, computing and information science, and applied mathematics, this book describes the emerging field of study at the interface of these areas. You can download the complete pre-publication draft.
• Rensselaer Team Shows How To Analyze Raw Government Data - Nov 19, 2010.
  Which White House staffer has the most visitors? How do smoking quit rates, state by state, relate to unemployment, taxes, and violent crimes? RPI team finds the answers to these questions and more
• Beginner's Guide To Web Data Analysis: Ten Steps To Love & Success - Nov 18, 2010.
  What if I was given the login and password to someone's web analytics data and asked to "find something interesting?" How would I start the process of web data analysis right?
• Digital Keys for Unlocking the Humanities Riches - Nov 16, 2010.
  The next big idea in language, history and the arts? Data.
• Massive Scale Data Mining for Education - Nov 16, 2010.
  in the near future, if millions of students start learning math using online computer software, we will have massive new data about how students work. There will be paths of learning in the data, some of which quickly reach mastery, others of which go off in the weeds.
• Why Are Business Analytics Important - Nov 15, 2010.
  You're probably worse at analyzing data than you think. That's the conclusion of wide-ranging illusory superiority research carried out by psychologists.
• Customer-Base Analysis Using Repeated Cross-Sectional Summary (RCSS) Data - Nov 14, 2010.
  Can customer data be stored and analyzed in an easy-to-manage, scalable, privacy-preserving format without significantly compromising the inferences that can be made about the customers' purchasing activity?

News Briefs

• Insurers Test Data Profiles to Identify Risky Clients - Nov 22, 2010.
  Life insurers are testing an intensely personal new use for the vast dossiers of data being amassed about Americans: predicting people's longevity.
• Congress Discussing Do Not Track Registry - Nov 21, 2010.
  The US Congress is discussing the possibility of instituting a registry for people who do not want websites to track them,
• Gravity knows what you like; soon advertisers will, too - Nov 21, 2010.
  CEO Amit Kapur says Gravity knows what you like and what you want better than any search engine. That information, of course, is gold to advertisers.
• VC view on Mobile Data Analytics - Nov 19, 2010.
  new smartphones capture more interesting data which cab drive the creation of a new and interesting analytics. As a result, I am becoming interested in mobile data analytics and have been actively looking for investment opportunities in this sector.
• UC Riverside receives Gates Foundation grant to fight malaria - Nov 15, 2010.
  Grant for innovative health research project by Eamonn Keogh, on Counting and classifying insects with ultra-cheap sensors.
• Privacy Advocates Call Out Google, Facebook and Other Data Miners - Nov 13, 2010.
  Industry coalition is pushing for tougher privacy legislation - but is it a trojan source to sneak in monitoring? Congress is currently considering legislation to make it easier to opt out of data mining.
• Stage Set for Showdown on Online Privacy - Nov 12, 2010.
  FTC exploring a stricter standard that requires a "do not track" option on a Web site or browser similar to the "do not call" lists, while Commerce officials favors letting the industry regulate itself with companies posting their privacy policies online.
• Aster Data Findings From Big Data Summit Conferences - Nov 12, 2010.
  including Discover New Market Opportunities (30% of respondents), Behavioral Targeting/Advertising (16%), Social Network Analysis (SNA): (15%), and more.
• Data marketplace Infochimps raises $1.2M to build out its data offerings - Nov 10, 2010.
  The company's focus has been on gathering social media and Web analytics, as well as government data.
• Patent: Data mining of unfiltered controller data - Nov 10, 2010.
  A data analysis system for utilization in an industrial environment comprises an aggregation component that aggregates data from multiple controllers on a factory floor.
• International Institute for Analytics thrives as a peer-generated research community - Nov 10, 2010.
  Members lead research agenda to advance the analytics profession and their careers

CFP - Calls for Papers (see also All CFP)

• Advances in Social Network Analysis and Mining, Workshop proposals, due Dec 1
• INFORMS Prize: best analytics/OR organization, due Dec 11
• PReMI'11: Pattern Recognition and Machine Intelligence, due Dec 15
• ICML 2011 Workshop Proposals, due Jan 14
• ICML 2011 Tutorial Proposals, due Jan 14
• ICWSM 2011: Weblogs and Social Media, due Jan 31
• KDD-2011: Knowledge Discovery and Data Mining Workshop proposals, due Feb 1
• ISMIS 2011, due Feb 7
• KDD-2011: Knowledge Discovery and Data Mining, due Feb 11
• UAI 2011, due Mar 18
• IDA 2011, due May 14

Quote

Uncertainty is the only certainty there is, and knowing how to live with insecurity is the only security.

John Allen Paulos