Improving your skills as a data scientist can be facilitated by reviewing the work of other professionals in the field. Having Kaggle as a resource means you can closely examine winning solutions of past machine learning competitions. Here is a lecture by Phil Brierley, a 3 time Kaggle winner and the “P” in POWERDOT, the winning team of the now completed Heritage Health Prize. Phil’s team received a check for $500,000.
Phil’s presentation was at the MelbURN R User Group discussing a few techniques to improve the accuracy of your predictive models and also a bit about the Heritage Health Prize story and how his team’s win unfolded. The video of the talk is below and the R code used in the simulations can be found HERE.
By education, Phil Brierley is a Mechanical Engineer and became involved in predictive analytics during his doctorate where he developed intelligent control systems using neural networks. Phil is the owner of Tiberius Data Mining. He has worked at NAB and IBM in advanced analytics and is also freelance data scientist currently working for a Hedge Fund.