Knowledge discovery in biological big data using near unsupervised learning: Keynote presentation

Abstract

Unsupervised learning is used for analysing and clustering data when the expected cluster labels are completely absent. When we know only a little about the data labels, i.e., class labels are scarce, but available for a small amount of data, it is still challenging to make conclusions, although this may be the case for many real world data mining problems. We name the type of learning algorithms useful in this scenario as Near Unsupervised Learning (NUL). My group has been developing NUL algorithms over a period of 14 years [1,7, 11] and some of these developments are based on Growing Self Organising Maps [9–11]. The concept and the algorithm development in NUL and the application in variou..