An early adopter of MOOCs, Steve Wells changed his 20+ year relationship with data after taking Andrew Ng’s course on Machine Learning in 2012. He now professes that probability isn’t just a theory. He takes a firm stance that even little data deserves the right to be normal.
For Steve, being mean is just an average day. For instance, he enjoys pointing out the gambler’s fallacy to players at the roulette table. He duly admits that though events in his life are assuredly correlated, the causation is usually his own fault, while leaning heavily on confounding variables when it’s not.
He wishes that data mining came with a headlamp while at the same time recognizing that forecasting and chaos theory can make for some fascinating mistakes. He displays intervals of confidence punctuated by systemic bias and though considered creative by some, his usual hypothesis is null.
As a consultant he’s keen to point out the importance of backing up your lies with statistics and that statistical inference is the art of proving a conjecture. As a sample size of one, none of this is statistically significant nor representative and he aims to keep it that way.
Data-Driven Documents: D3Beer Consumption immediately following Prohibition (Udacity)
Hunt him down at any of the following resources
...and he's always happy to connect with others
on LinkedIn so send in your request...