Soft Computing and Industry Recent Applications, R. Roy, M. Koppen, S. Ovaska, T. Furuhashi and F. Hoffmann, editors, pages 597–608, Springer-Verlag, 2002. Genetic Programming for Combining Neural Networks for Drug Discovery
Download:
pdf
by W. B. Langdon, S. J. Barrett, B. F. Buxton
ftp://cs.ucl.ac.uk/genetic/papers/WBL_wsc6.pdf
Add To MetaCart
Abstract:
Abstract. We have previously shown on a range of benchmarks [Langdon and Buxton, 2001b] genetic programming (GP) can automatically fuse given classifiers of diverse types to produce a combined classifier whose Receiver Operating Characteristics (ROC) are better than [Scott et al., 1998]’s “Maximum Realisable Receiver Operating Characteristics” (MRROC). I.e. better than their convex hull. Here our technique is used in a blind trial where artificial neural networks are trained by Clementine on P450 pharmaceutical data. Using just the networks, GP automatically evolves a composite classifier. 1
Citations
| 1782 | Genetic Programming: On the Programming of Computers by Means of Natural Selection Cambridge – Koza - 1992 |
| 569 | Adaptive mixtures of local experts – Jacobs, Jordan, et al. - 1991 |
| 18 | Multiple interacting programs: A representation for evolving complex behaviors – Angeline - 1998 |
| 18 | Genetic programming for combining classifiers – Langdon, Buxton - 2001 |
| 10 | Evolving receiver operating characteristics for data fusion – Langdon, Buxton, et al. - 2001 |
