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

<div>Mean number of chosen features and stability of the 6,997 configurations for data set <span class="monospace">AP_Breast_Ovary</span> whose mean misclassification rate is within 0.05 of the best classification accuracy.</div>

Computational and Mathematical Methods in Medicine

fig8

Figure 8

Figure 8: A Multicriteria Approach to Find Predictive and Sparse Models with Stable Feature Selection for High-Dimensional Data 