Online learning isotonic separation for the study of largescale data
Ashthika Balakrishnan · 2018
Isotonic Separation (ISs) [2] is a linear programming technique which separates data based on domain knowledge and isotonic consistency condition. It is a supervised machine learning [3] technique in which an isotonic function obtained from training set is represented as a Linear Programming Problem (LPP) with an objective of minimizing the number of misclassifications. In this paper, we focus on adapting Isotonic Separation as an online learning process. The proposed method consists of two components: the learning prototypes (LPs) and the learning isotonic separators (LISs). The proposed method has been compared with batch learning Isotonic Separation.