Skip to main content

Table 5 Statistics about the comparative performance of the base k-NN classifiers and their label similarity-incorporated versions using information in the whole GO biological process ontology for all as well as small (size ≤ 30) target classes in Mnaimneh et al's data set, measured in terms of the number of classes for which AUC scores are improved by the latter over the former, and the average and maximum improvement in AUC scores over all classes.

From: Incorporating functional inter-relationships into protein function prediction algorithms

  Total # classes # Classes improved Average improvement over all classes Maximum improvement
All classes 137 71 0.0167 (2.65%) 0.2492 (53.93%)
Small classes 47 29 0.0363 (6.32%) 0.2492 (53.93%)