Unsupervised Instance Selection with Low-Label, Supervised Learning for Outlier Detection

TL;DR

This paper introduces UNISEL, an unsupervised instance selection method, which, when combined with supervised learning, effectively detects outliers with minimal labeled data, outperforming traditional active learning in certain scenarios.

Contribution

The paper proposes UNISEL, an unsupervised instance selection technique that enhances outlier detection efficiency, especially under low-label conditions, and demonstrates its effectiveness compared to active learning.

Findings

UNISEL performs comparably to active learning with fewer labels.

Combining UNISEL with active learning yields superior outlier detection results.

UNISEL offers practical time savings and better generalizability.

Abstract

The laborious process of labeling data often bottlenecks projects that aim to leverage the power of supervised machine learning. Active Learning (AL) has been established as a technique to ameliorate this condition through an iterative framework that queries a human annotator for labels of instances with the most uncertain class assignment. Via this mechanism, AL produces a binary classifier trained on less labeled data but with little, if any, loss in predictive performance. Despite its advantages, AL can have difficulty with class-imbalanced datasets and results in an inefficient labeling process. To address these drawbacks, we investigate our unsupervised instance selection (UNISEL) technique followed by a Random Forest (RF) classifier on 10 outlier detection datasets under low-label conditions. These results are compared to AL performed on the same datasets. Further, we investigate the combination of UNISEL and AL. Results indicate that UNISEL followed by an RF performs comparably to AL with an RF and that the combination of UNISEL and AL demonstrates superior performance. The practical implications of these findings in terms of time savings and generalizability afforded by UNISEL are discussed.