Regularized F-Measure Maximization for FeatureSelection and Classification

[摘要] Receiver Operating Characteristic (ROC) analysis is a common tool forassessing the performance of various classifications. It gained much popularity in medical and other fields including biological markers and, diagnostic test. This is particularly due to the fact that in real-world problemsmisclassification costs are not known, and thus, ROC curve and related utilityfunctions such as F-measure can be more meaningful performance measures.F-measure combines recall and precision into a global measure. In this paper, we propose a novel method through regularized F-measure maximization.The proposed method assigns different costs to positive and negative samples and does simultaneous feature selection and prediction withL1penalty. This method is useful especially when data set is highly unbalanced, or thelabels for negative (positive) samples are missing. Our experiments with thebenchmark, methylation, and high dimensional microarray data show that the performance of proposed algorithm is better or equivalent compared with the other popular classifiers in limited experiments.

[学科分类] 基础医学