While feature selection is almost a solved problem in data science, feature engineering is still quite a mystery. In this talk I will outline a method that I use to solve feature engineering, with a goal to provide a generalized framework to tackle both feature engineering and selection simultaneoously.

Outline

The first few slides will talk about the application of genetic algorithms (GA) to feature selection. The next couple of slides will talk about advancements made to GAs by use of a multi-dimensional covariance map, a method that I developed. The next couple of slides will talk about genetic programming (GP) and how one can use the multi-dimensional covariance map to augment the convergence of GPs.

Requirements

A good understanding of machine learning fundamentals

Speaker bio

I’m currently a principal data scientist at Intuit. A public but slightly dated bio is available here: https://www.analyticsvidhya.com/datahack-summit-2018/speakers/sidharth-kumar/ An informal writeup on me is available here: http://humansofanalytics.com/stories/sidharth-kumar-data-science-savant-machine-learning-aficionado-and-ardent-chess-player/