Figure 5
From: A fuzzy rank-based ensemble of CNN models for classification of cervical cytology
The non-linear functions used to generate fuzzy ranks in the proposed ensemble framework. x denotes the probability of a class of a sample data. (a) Quantifies the deviation from its objective for a class having prediction probability. Deviation decreases when x decreases. Eventually it becomes 0 when \(\hbox {x} = 1\). (b) Quantifies the reward to be given to a class having prediction probability x. Reward increases when x increases. Eventually it becomes 1 when \(\hbox {x} = 1\).
