A proposal for PU classification under Non-SCAR using clustering and logistic model

TL;DR

This study introduces a simple clustering and logistic regression algorithm for positive-unlabeled classification without the SCAR assumption, demonstrating effectiveness and robustness through real and synthetic data tests.

Contribution

The paper proposes a novel, computationally simple PU classification method that works under non-SCAR conditions and assesses its robustness against perturbations.

Findings

Clustering-based cleaning labels improve PU classification when SCAR is violated.

The proposed method performs well on 11 real datasets and synthetic data.

LassoJoint shows moderate robustness to SCAR condition perturbations.

Abstract

The present study aims to investigate a cluster cleaning algorithm that is both computationally simple and capable of solving the PU classification when the SCAR condition is unsatisfied. A secondary objective of this study is to determine the robustness of the LassoJoint method to perturbations of the SCAR condition. In the first step of our algorithm, we obtain cleaning labels from 2-means clustering. Subsequently, we perform logistic regression on the cleaned data, assigning positive labels from the cleaning algorithm with additional true positive observations. The remaining observations are assigned the negative label. The proposed algorithm is evaluated by comparing 11 real data sets from machine learning repositories and a synthetic set. The findings obtained from this study demonstrate the efficacy of the clustering algorithm in scenarios where the SCAR condition is violated and further underscore the moderate robustness of the LassoJoint algorithm in this context.