KDnuggets™ News 11:n24, Oct 12

Features

Columbus day to celebrate and reflect upon discovery in cyberspace? - Oct 10, 2011.
an appropriate day to celebrate and reflect upon DISCOVERY, an essential part of our field. What can we learn from mistakes of Christopher Columbus that applies to discovery in cyberspace?
How Obama's data-crunching prowess may get him re-elected - Oct 11, 2011.
So far in the presidential election of 2012, there is only one campaign that is doing cutting-edge work with data. (this CNN story features KDnuggets)
Judge forbids use of Bayes Theorem. Really? - Oct 9, 2011.
A UK judge rules ruled that a Bayes theorem can no longer be used in court. This could affect thousands of cases. Seriously.
1 week until Predictive Analytics World October 2011 in New York - Oct 11, 2011.
Register by Friday Oct 14 and Save $200. Can't Miss Events: Everyday Analytics Keynote with Tom Davenport; Building Watson - Deep QA Project; The Biggest Predictive Analytics Competitions, and more.
Survey: Predictive Analytics in the Cloud - Sep 30, 2011.
voice your opinion on how predictive analytics and the cloud are impacting your business now and in the future. Get access to results with registration.
KDD-2012 Call for Papers, due Feb 10, 2012 - Oct 6, 2011.
18th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, August 12-16, 2012, Beijing, China - the premier conference in the field.
Top news for Oct 2-8 - Oct 9, 2011.
Data Mining Ontology Foundry; Data Mining Research Notes; KDD-12
Top jobs: Data Mining Scientist at Apple; Research Scientist - Knowledge Representation and Reasoning at IBM Research, Dublin.
Top news for Sep 25 - Oct 1 - Oct 2, 2011.
Poll Results: How different is "Data Science" from "Statistics"?; Data Scientist Summit 2011 Videos;
Top jobs: Research Scientist - ML / DM / Learning Systems at IBM Research, Dublin, Ireland; DM Specialist at Statistics Solutions, Palm Harbor, FL.
Top news, jobs in September - Oct 1, 2011.
New Poll: Is Data Science different from Statistics; Data Scientist Summit Videos;
Top jobs: Researcher - Recommender Sys/Analytics at Ericsson, Ireland; Fraud Investigation at MasterCard.
Additions to KDnuggets Directory in September - Oct 1, 2011.
New Companies, Datasets, Education, Meetings, Software, Blogs

Courses, Events

Learn What Really Works in Data Mining: Nov in Dallas | Dec in Los Angeles - Oct 11, 2011.
insist on 1000% ROI from your data mining initiatives! Learn how to get it from TMA Predictive Analytics and Data Mining course series.
Sentiment Analysis Symposium Highlights, San Francisco, Nov 9, 2011 - Oct 4, 2011.
Zynga, TripAdvisor, HP, Red Cross to Headline the Symposium. 3rd Symposium highlights role sentiment plays in social and media measurement, customer experience, and market research.

Webcasts

FICO Webinar, Oct 19: Managing Predictive Models in a Regulated, Dynamic Economy - Oct 6, 2011.
focuses on best practices to build, improve, and refine methodologies for achieving regulatory compliance while simultaneously spurring growth. FICO is presenting the webinar twice, in order to focus separately on North American and European compliance issues.
Slides and replay from "R and Hadoop" webinar - Oct 3, 2011.
Over 1000 people signed up for a record-setting Revolution Analytics webinar.

Software

11Ants Customer Churn Analyzer beats 85% of submissions in KDD Cup 2009 - Oct 6, 2011.
11Ants Analytics Customer Churn Analyzer beat 85% of submissions in KDD Cup 2009 - top international data mining contest - with only 55 minutes of human work involved.
Visualize US Census inside Bing Maps - Oct 11, 2011.
CensusViewer combines a blazingly fast patent-pending visualization technology with the census data and Bing Maps. It includes the Census 2000 and Census 2010 Race and Latino data sets with more than 100 million records each.
Data mining with Ruby and Twitter - Oct 10, 2011.
great examples of using Ruby to get Twitter user data, measure friend popularity, analyze followers by location, building a bar chart of tweet days, and more.
Rice Analytics Issued Fundamental Patent on RELR Method - Oct 7, 2011.
US Patent 8,032,473 covers important error modeling and dimension reduction aspects of RELR.
E.U. Sets 2013 Deadline for Open Source Public Data Mining Portal - Oct 4, 2011.
Making good use of public data can make your life better; Research in genomics, pharmacology or the fight against cancer increasingly depends on the availability and sophisticated analysis of large data sets.
DMO (Data Mining Ontology) Foundry - Oct 2, 2011.
will gather the most significant ontologies concerning data mining and the different algorithms and resources for the knowledge discovery process.
MariaDB: the new MySQL? Interview with MySQL main author - Sep 29, 2011.
I want to ensure that the MySQL code base (under the name of MariaDB) will survive as open source, in spite of what Oracle may do.

Jobs

Data Mining Scientist at Apple. Inc., Austin, TX - Oct 5, 2011.
designing, developing, and fielding data mining solutions that have direct and measurable impact to Apple; work with business managers and executives to help identify viable data mining opportunities and then implement end to end analytical solutions.
Research Scientist - Knowledge Representation and Reasoning at IBM Research, Dublin, Ireland - Oct 4, 2011.
researching novel representations and algorithms to enable decision makers to exploit urban and environmental data and information and linked open data
(Senior) Director Analytics at Innovation Labs, 24/7 Customer, Bangalore, India - Sep 30, 2011.
Deliver Predictive Analytics, Data Mining, Text Mining, Machine Learning driven products and solutions to 24/7 Clients.
Associate Predictive Modeler at Allstate Insurance, Menlo Park, CA - Sep 28, 2011.
work with a team of highly skilled and expert predictive modelers at one of the leading property and casualty insurance companies.
Senior Quantitative Engineer at casual games company, Manhattan, New York - Sep 28, 2011.
We are the world's leading multi-platform casual games company, with expertise in the online, mobile, interactive TV and retail categories. We need you to help us tap the potential power of our vast and diverse data.

Academic/Research positions

Lecturer in Computing & Informatics (Business Intelligence and Data Mining) at Bournemouth U., UK - Oct 10, 2011.
make a significant contribution towards our research capacity in the areas of business intelligence, data mining and complex software systems.
Senior Lecturer in Predictive Analytics and Data Mining at Bournemouth U., UK - Oct 10, 2011.
play a significant role as a part of the coordinating team in an exciting, large European project on Computational Intelligence Platform for Evolving and Robust Predictive Systems (INFER)
4 PhD Studentships in Predictive Modelling, Learning, Design at Bournemouth U., UK - Oct 10, 2011.
Four 3 year, fully funded PhD research studentships are available to work on a range of exciting research areas within the framework of a large EU funded INFER project.
PhD Student Position in Machine Learning, Data analysis at Aalto U., Helsinki, Finland - Oct 7, 2011.
develop advanced machine learning methods for nonlinear dimensionality reduction, visualization, and exploratory data analysis with multiple data sources. Deadline: Oct 31, 2011
Postdoctoral Researcher in Analysis and Modeling of Social Networks at Indiana U., Bloomington, IN - Oct 4, 2011.
work on analysis and modeling of social media data, including information diffusion patterns, epidemic models for the spread of ideas, interactions between network traffic and structure dynamics, and agent-based models.
Research Fellowship in Opinion Mining at INESC-ID, Lisbon, Portugal - Sep 30, 2011.
developing ML techniques for classifying text form different genres, including news, blogs and micro-blogs, mining opinions from social media about news topics and personalities, and detecting information flow.
Post-doctoral opening in Healthcare Analytics at U. of Notre Dame, Notre Dame, IN - Sep 28, 2011.
help build a centralized database of medical records of rare disease patients; responsible for designing the EMR and also for advanced analytics for better patient management and outcome.

Competitions

Cleveland Clinic: Build an Efficient Pipeline to Find the Most Powerful Predictors - Oct 8, 2011.
Given ~7400 genetic variables and outcome data, deliver R package that selects the most powerfully predictive variables, then uses these to accurately predict outcomes. 30K prize from the Cleveland Clinic.
Give Me Some Credit - Kaggle Competition - Sep 29, 2011.
improve the state of the art in credit scoring, by predicting the probability that somebody will experience financial distress in the next two years.
Heritage Health Prize $3M competition - 6-month milestone winners - Sep 28, 2011.
a report on the 6-month milestone awards in Heritage Health Prize $3M competition.

Meetings

Text West: Early Bird ends Friday - Oct 11, 2011.
Text West (Nov 10-11, San Jose) provides an excellent opportunity to network with new vendors and end users on the other side of the country.
Big Data Impressions from Strata Summit NYC - Oct 5, 2011.
a report on the Strata Summit on the Business of Data
KDD-2012: August 12-16, Beijing, China - Sep 30, 2011.
the premier international forum for data mining researchers and practitioners from academia, industry, and government to share their ideas, research results and experiences.

Publications

Music Data Mining Research and Bibliograhy - Oct 9, 2011.
a very comprehensive survey of Music data mining research, including papers and datasets for Association Mining, Genre Classification, Rhythm Classification, Artist Classification, Mood Detection and Classification and more.
Data Science: a literature review - Oct 8, 2011.
Is it a useful rebranding of analytics or a statistician who lives in select regions of california or works at bitly?
Process Mining Manifesto from IEEE Task Force on Process Mining - Oct 7, 2011.
The goal is to increase the visibility of process mining as a new tool to improve the (re)design, control, and support of operational business processes.
Data Mining Research Notes - Oct 1, 2011.
a categorized list of Wikipedia articles useful for data mining students, covering many important concepts and how they relate to each other.
The Power of Text Analytics: Developments, Insights, and Innovation - Sep 29, 2011.
3 pioneers in text analytics and big data talk about their personal experiences in the analytics industry and the future outlook of the text analytics market.
Dancing with the Stats - Sep 29, 2011.
Is it possible to predict the finalists on Dancing with the Stars before the contestants ever step on the floor? Surprisingly, beauty, fame level, and ethnicity seem to matter very little.

News Briefs

Angoss Predictive Analytics Suite 7.6 - Oct 10, 2011.
Features in-database analytics, market basket analysis and advanced data visualization enhancements.
Oracle launches Exalytics, an appliance for big data - Oct 8, 2011.
Exalytics machine not only handles relational data, not only multidimensional data, [but] also analyzes unstructured data at the speed of thought. Nothing is faster, according to Oracle CEO Larry Ellison.
Machine learning answers: Get in or out of the stock market? - Oct 6, 2011.
A high-tech startup has developed a computerized machine learning technique to predict the stock market. It now answers the biggest investment question.
Revolution Analytics partners with Cloudera for Big Analytics Solution - Oct 5, 2011.
New Solution Integrates R with Cloudera's Distribution Including Apache Hadoop to Deliver Powerful Solution to Tame Big Data
Predixion Software Raises $6 Million in Series B Financing - Oct 4, 2011.
will use the funding proceeds to build out additional healthcare industry solutions, further expand its product development activities and scale its service, sales and marketing organizations
Ingres Becomes Actian To Take Action on Big Data - Oct 3, 2011.
new strategy to enable companies to develop Action Apps, lightweight, consumer-style applications that automate business actions triggered by real-time changes in data.
Pittsburgh Pirates tap SAS(r) Analytics - Oct 3, 2011.
Baseball loves the numbers, both traditional RBIs and ERAs plus newer sabermetric measures like walks plus hits per inning pitched (WHIP) and value over replacement player (VORP).
How Benford's law could have prevented the Greek disaster - Sep 28, 2011.
Analysis of official statistics of the EU member states for the last 11 years clearly shows that Greek financial data was the most suspect.

CFP - Calls for Papers

PAW-SF-12: Predictive Analytics World, due Oct 13
PAW-Toronto-12: Predictive Analytics World, due Oct 13
TAW-SF-12: Text Analytics World, due Oct 13
SBP12: Social Computing, Behavioral-Cultural Modeling, and Prediction, due Nov 4
Salford DM 2012: Salford Data Mining Conference 2012, due Nov 4
3Clust: Multi-view data, High-dimensionality, External Knowledge: a Unified Approach to Clustering, due Jan 13
PAW-Chicago-12: Predictive Analytics World, due Jan 19
KDD-2012: 18th ACM SIGKDD Conf. on Knowledge Discovery and Data Mining, due Feb 10

Quote

John Kay muses on interpreting statistical data: Always ask of such data "what is the question to which this number is the answer?". "Earnings before interest, tax, depreciation and amortisation on a like-for-like basis before allowance for exceptional restructuring costs" is the answer to the question "what is the highest profit number we can present without attracting flat disbelief?".