Skip to main content

Adviser: Learning How to get A Second Opinion on Your Analysis when it's Important to get it Right

Leland Wilkinson (Skytree)
Data Science
GA Ballroom J
Tutorial Please note: to attend, your registration must include Tutorials on Tuesday.
Average rating: ***..
(3.50, 2 ratings)

Skytree Adviser is an expert system for machine learning. It not only chooses appropriate models, but it also evaluates them with unique algorithms for detecting miss-specifications, outliers, and other anomalies. It was developed through iterative testing on two different populations: novices (marketers, executives, students…) and experts (Ph.D. statisticians, computer scientists, analysts, …). Novices have been able to fit specialized models, such as ordinal logistic regression or spectral clustering, and experts have discovered anomalies in their model-fitting process that they had not noticed when using standard ML platforms or statistics packages. This unusual range of target users was made possible by the development of a uniquely simple interface (called a “button tree”) and a rich output format (an HTML document containing an introduction, tables, dynamic graphics, technical references and numerous links to tutorial Web sites). This tutorial will feature real data examples to illustrate the unique detective work that Adviser provides for discovering patterns in data that most analysts and programs tend to miss. Because of the flexibility of the Adviser design, participants ranging from Excel users to advanced analysts can benefit from this tutorial.

Photo of Leland Wilkinson

Leland Wilkinson

VP of Data Visualization, Skytree

Leland Wilkinson is currently adjunct professor of computer science at the University of Illinois at Chicago. Previously, he was adjunct professor of statistics at Northwestern University and President of SYSTAT Inc., a statistical software company he founded in 1984. Wilkinson is a fellow of the American Statistical Association, an elected member of the International Statistical Institute, and a fellow of the American Association for the Advancement of Science. He was vice-chair of the board of the National Institute of Statistical Sciences and a member of the Committee on Applied and Theoretical Statistics at the National Academy of Sciences. Wilkinson recently served on the NAS Panel on Developing Science, Technology, and Innovation Indicators for the Future. His projects have included books, journal articles, the original SYSTAT statistics package, numerous open-source software packages, and patents in visualization and distributed computing. He has an A.B. from Harvard University, an S.T.B. degree from Harvard Divinity School, and a Ph.D. from Yale.