Latest KDnuggets News 10:n06, (March 24, 2010) Features:
- New Poll: Is anonymization of large datasets still possible?
- Poll Results: Sports Analytics Are Useful
- KDD Cup 2010: Educational Data Mining
- KDD-2010 Workshops - calls for papers
- Next Neflix Prize cancelled due to privacy concerns
- Most viewed items for week Mar 14-20
- Most viewed items for week Mar 7-13
- Predictive Analytics World: Save-the-Date and Call-for-Speakers
Quote
Empirical evidence is that 80-90% of the claims made by epidemiologists are false; their claims do not replicate when retested under rigorous conditions. The net effect of ignoring multiple testing is to exploit randomness.
S. Stanley Young, National Institute of Statistical Sciences