Skip to main content
. 2009 Aug 27;10(Suppl 8):S4. doi: 10.1186/1471-2105-10-S8-S4

Table 6.

Overall F1 measure of the entire classification. The global F1 measure of the classification problem is computed by two different types of averages: micro-average and macro-average. In micro-averaging, the F1 measure is calculated globally over all categories, while in macro-averaging, F1 is computed locally over each category first, and then the average over all categories is taken.

MAN FEAT
F1(micro-averaged) 0.56 0.55
F1(macro-averaged) 0.43 0.19