Table 7.

Classification performance on the Olive Oil data for the variable selection algorithm with updating and for previous analyses of these data. Mean classification performance for the 50 random splits of the data are reported with standard deviations in parentheses. For the Variable Selection Only results, the maximum number of selected wavelengths was restricted to be six to avoid degeneracies.

Method	Misclassification Rate
Variable Selection and Updating	6.9% (5.4)
Variable Selection (Greedy) and Updating	16.6% (11.3)
Variable Selection Only	17.9% (10.9)
Dean et al. (2006)	11.9% (6.3)
Downey et al. (2003)	6.1–19.0%
Transductive SVMs	12.4% (7.5)
Random Forests	19.3% (6.5)
AdaBoost.M1	34.1% (9.3)
Bayesian Multinomial Regression	57.0% (1.2)