Date and time: December 15 2005 (10:30 – 12:00).
Room: 336 RB Non–standard venue!
Automating the 4FT–Miner Set–Up: Two Example Approaches from the Toolbox of AI
- Sebastian Schutte, University of Freiburg, Germany
The LISp–Miner environment provides a challenging field for the development of powerful and easy–to–use data mining techniques. There are two main directions in the ongoing efforts to extend the software. Simplifying the handling and extending the functionality are the main directions of development. This talk gives a brief overview of the ongoing activities and focusses afterwards on two ways how to simplify the setup. A rule based system for using context information for the 4ft–Miner setup is presented and the outlines of a genetic algorithm are discussed, that optimises the setup toward a user–defined fitness function.
Downloads: slides 1