Augmenting Multi-Instance Multilabel Learning with Sparse Bayesian Models for Skin Biopsy Image Analysis

Table 3

Evaluation criteria for multilabel learning.

Name	Equation
ℎ𝑙𝑜𝑠𝑠	Evaluate the number of misclassified label pairs
one-error	Evaluate the portion that a label of highest probability is not a correct label
𝑐𝑜𝑣𝑒𝑟𝑎𝑔𝑒	Evaluate the average distance to go down to find the proper label for a given image
𝑟𝑙𝑜𝑠𝑠	Evaluate the average fraction of label pair that are misordered in the ranking list