【論】Guyon,2002,Gene selection for cancer class～

I.Guyon, J.Weston, S.Barnhill, and V.Vapnik.
Gene selection for cancer classification using support vector machines.
Machine Learning, Vol.46, pp.389--422, 2002
[PDF][Web Site]

・SVMのマイクロアレイデータ解析への応用の代表的な論文。
・データ
その1 Differentiation of two types of Leukemia, 72サンプル×7129遺伝子[Golub]
その2 Colon cancer diagnosis, 62サンプル×2000遺伝子[Alon]
・クラス分け法
1.SVM RFE
2.Linear Discriminant Analysis RFE
3.Mean Squared Error (Pseudo-inverse) RFE
4.Baseline method
・クラス分け評価法：Leave-one-out success rate

・概要「In this paper we investigate pruning techniques that eliminate some of the original input features and retain a minimum subset of features that yield best classification performance.」
・従来法「Each coefficient w_i is computed with information about a single feature (gene) and does not take into account mutual information between features.」
・結果「All our feature selection experiments using various classifiers (SVM, LDA, MSE) indicated that better features are obtained by using RFE than by using the weights of a single classifier. Similarly, better results are obtained by eliminating one feature at a time than by eliminating chunks of features. However, there are only significant differences for the smaller subset of genes (less than 100).」
・展望「We envision that linear classifiers are going to continue to play an important role in the analysis of DNA micro-array because of the large ratio number of features over number of training patterns.」
・Golubが使用した相関係数　w_i = (μ_i(+) - μ_i(-))/(σ_i(+) + σ_i(-))
・Fisher's discriminant criterion (μ_i(+) - μ_i(-))²/(σ_i(+)² + σ_i(-)²)