Local Clustering Conformal Predictor for Imbalanced Data Classification

Abstract

Abstract. The recently developed Conformal Predictor (CP) can provide calibrated confidence for prediction which is out of the traditional predictors' capacity. However, CP works for balanced data and fails in the case of imbalanced data. To handle this problem, Local Clustering Conformal Predictor (LCCP) which plugs a two-level partition into the framework of CP is proposed. In the first-level partition, the whole imbalanced training dataset is partitioned into some class-taxonomy data subsets. Secondly, the majority class examples proceed to be partitioned into some cluster-taxonomy data subsets by clustering method. To predict a new instance, LCCP selects the nearest cluster, incorporated with the minority class examples, to build a re-balanced training data. The designed LCCP model aims to not only provide valid confidence for prediction, but significantly improve the prediction efficiency as well. The experimental results show that LCCP model presents superiority than CP model for imbalanced data classification.

A survey on confidence calibration of deep learning under class imbalance data

“…Research Status: Conformal prediction. Considering that the original conformal prediction is not well applicable to class imbalance data, Wang et al [157] proposed local clustering conformal predictor, which plugs a two-level partition into the conformal prediction framework, wherein the entire imbalanced training dataset is divided into some classtaxonomy data subsets at the first level, and the majority class examples are further divided into some cluster-taxonomy data subsets at the second level by using a clustering technique. Based on similar motivations, Sun et al [158] proposed Mondrian cross-conformal prediction to improve confidence under imbalance bioactivity datasets, which combines the Mondrian inductive conformal prediction [159] with cross-fold calibration.…”