ClassDist u j,c k ) may also be any aggregation function computed on the individual pair distances between one unlabeled instance u j ∈U i and every labeled instance from class c K ∈C i known at iteration i at the current implementation we are using the median.
Typically, the class imbalance problem occurs when there are significantly more instances from one class relative to other classes.
When computing first-hit for a given class we have excluded the experiments where the labeled set for the first iteration contains instances from that class.
In fact, the real dataset in this study is imbalanced, and the instances from negative class take the majority of the data.
The training data is partitioned into sub-samples with each sub-sample containing an equal number of instances from each class, except for last sub-sample (in some cases).
Resampling datasets in order to reach balanced distributions is a common practice that sometimes improves classification performance, as the model encounters an equal number of instances from each class, thereby producing a more appropriate discriminative function as opposed to a function obtained from skewed distributions.
At every partition a test is performed for every attribute and the test set that maximizes the entropy-based gain ratio [ 78] is selected, leading to a tree where every leaf contains instances from one class when there is no over-fitting.
Moreover, when in the presence of an imbalanced class distribution, getting labeled instances from minority classes might be very costly, requiring extensive labeling, if queries are randomly selected.
On a highly imbalanced class distribution, it is particularly demanding to identify instances from minority classes.
The second method used is upsampling, in which instances from the classes with proportionally low numbers of instances are duplicated to reach a more balanced class distribution.
