Learning with Constrained and Unlabelled Data
CVPR '05 Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Volume 1 - Volume 01
Statistical Comparisons of Classifiers over Multiple Data Sets
The Journal of Machine Learning Research
Semi-Supervised Learning
Hi-index | 0.00 |
The analysis of cancer gene expression is intrinsically a semi-supervised problem, as one is interested in building a classifier for diagnosis, but also on finding new sub-classes of cancer. We propose here a method for Mixture Discriminant Analysis (MDA), which can simultaneously detect sub-classes of cancer and perform classification. We evaluate the method on 10 gene expression data sets. MDA not only improved the classification in some of these data sets, as it detected some known and putative sub-classes of cancer.