A Multicriteria Approach to Find Predictive and Sparse Models with Stable Feature Selection for High-Dimensional Data

Table 9

Pareto optimal configurations in Figure 10(b) (above the horizontal line) and accuracy optimal configurations (below the horizontal line). The ID column references Table 10.

ID	Classifier	Parameters	Filter	n.feats	Error	Error	Size	Size
					(train)	(test)	(train)	(test)
16	Lasso Log. Reg.		AUC	2730	0.215	0.466	2726.9	2726.1
17	Lasso Log. Reg.		MRMR	1972	0.219	0.491	1971.1	1970.2
18	GLM Boosting		MRMR	461	0.193	0.360	35.5	46.8
19	GLM Boosting		AUC	8330	0.221	0.494	11.8	14.0
20	GLM Boosting		Variance	8526	0.175	0.461	48.7	60.9