





















Abstract:In this paper, we consider the problem of conditional anomaly detection that aims to identify data instances with an unusual response or a class label. We develop a new non-parametric approach for conditional anomaly detection based on the soft harmonic solution, with which we estimate the confidence of the label to detect anomalous mislabeling. We further regularize the solution to avoid the detection of isolated examples and examples on the boundary of the distribution support. We demonstrate the efficacy of the proposed method on several synthetic and UCI ML datasets in detecting unusual labels when compared to several baseline approaches. We also evaluate the performance of our method on a real-world electronic health record dataset where we seek to identify unusual patient-management decisions.
| Comments: | Published at IEEE International Conference on Data Mining (ICDM 2011). https://doi.org/10.1109/ICDM.2011.40 |
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2604.21462 [cs.LG] |
| (or arXiv:2604.21462v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2604.21462 arXiv-issued DOI via DataCite (pending registration) |
|
| Journal reference: | IEEE International Conference on Data Mining (ICDM), pp. 735-743, 2011 |
| Related DOI: | https://doi.org/10.1109/ICDM.2011.40
DOI(s) linking to related resources |
From: Michal Valko [view email]
[v1]
Thu, 23 Apr 2026 09:17:03 UTC (376 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理,仅供阅读参考。 原文来自 — 版权归原作者所有。