Machine Learning Times
EXCLUSIVE HIGHLIGHTS
Incoherent AGI Hype Spurs An Industrywide Pivot To Hybrid AI
  Originally published in Forbes Recently on The Dr. Data Show,...
The AI Paradox: More Humanlike Means Less Autonomous
  Originally published in Forbes The AI executives are at...
How To Overcome The Confidence-Killer That Destroys Most Predictive AI Projects
  Originally published in Forbes When Henry Castellanos first presented...
You Must Address These 4 Concerns To Deploy Predictive AI
 Originally published in Forbes Most predictive AI projects fail to launch into production. The...

Archive for March, 2014

Applications of Predictive Modeling in Drug Development

 Recently, various empirical and semi-empirical models embedded in different modeling tools have been developed and recognized for their role in predicting pharmacokinetics of drugs in humans. These models have also been used to evaluate the effects of intrinsic (e.g., organ dysfunction, age, genetics) and extrinsic (e.g., drug-drug interactions) factors, alone or in combinations, on drug

Why Overfitting is More Dangerous than Just Poor Accuracy, Part II

 In part one, I described one problem with overfitting the data is that estimates of the target variable in regions without any training data can be unstable, whether those regions require the model to interpolate or extrapolate....

Proactively fighting fraud with data

 A recent SAP survey found that 41 percent of financial services organization felt that predictive analytics are more about minimizing risk than exploiting opportunities. The way that statistic is phrased, it sounds like the financial industry views...

The Future of Customer Service: Predictive, Personalized

 The future of customer service is scary and rewarding at the same time. It’s scary because the machine will know everything about you. It’s rewarding because shopping will become much easier as the machine makes more decisions...

Why Predictive Analytics Marketplaces are not taking off, and how to fix it

 Three main hurdles holding back Predictive Analytics Marketplaces are a highly fragmented data mining tools market, limited support for customization, and lack of commitment. We examine how to overcome them. An earlier article on KDnuggets noted that...

Page 2 of 2 1 2