Sigve Dreyer will, as part of his master of science work talk about Evolutionary Feature Selection. His work has been motivated by increasing data-dimensionality and the need to construct simple induction models. He will discuss his work on how to construct an abstract feature selection algorithm capable of using many different wrappers, and how it is tested on a wide range of datasets, revealing both it’s abilities and problems.
This lecture will take place in room 454, Friday February 28th at 12:00. The talk will last for about 30 min.