At JMP, we are fans of Professor David J. Hand, world-renowned, award-winning statistician, author and speaker. So in this latest episode of Statistically Speaking, we were delighted to feature his insightful plenary talk on "Assessing the Performance of Machine Learning Algorithms."
In this video clip, he considers, "How do we tell how good a classifier is?" and argues we need to consider this for a number of reasons.
The plenary was followed by a discussion with three exceptional statisticians:
- Diana Ballard, a statistician at Brewer Science, who has more than 20 years of manufacturing statistical experience.
- Jason Brinkley, a biostatistician, data scientist and health researcher who leads the Research Design and Analytics team in Abt Associates’ Data Science, Surveys and Enabling Technologies division.
- Jim Georges, who leads the Advanced Analytics Lab at SAS.
Their discussion was enlightening in so many ways. Some highlights from the discussion include:
- The importance of collecting high-quality data and performing good data preparation (because all data have issues).
- The value of really looking at the data.
- The crucial role of “statistical strategy” to best frame and solve the problem at hand.
As I reflect on Thanksgiving this week in the US, I am grateful for the statisticians in this episode of Statistically Speaking and all of the creative problem-solvers who use data visualization and statistics to discover insights and make better-informed decisions.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.