Conditional Anomaly Detection Using Soft Harmonic Functions: An Application to Clinical Alerting - Archive ouverte HAL Access content directly
Conference Papers Year : 2011

Conditional Anomaly Detection Using Soft Harmonic Functions: An Application to Clinical Alerting

(1) , (2) , (3) , (4) , (2)
1
2
3
4

Abstract

Timely detection of concerning events is an important problem in clinical practice. In this paper, we consider the problem of conditional anomaly detection that aims to identify data instances with an unusual response, such as the omission of an important lab test. 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 in detecting unusual labels on a real-world electronic health record dataset and compare it to several baseline approaches.
Fichier principal
Vignette du fichier
valko2011conditional.pdf (329.82 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-00642313 , version 1 (17-11-2011)

Identifiers

  • HAL Id : hal-00642313 , version 1

Cite

Michal Valko, Hamed Valizadegan, Branislav Kveton, Gregory Cooper, Milos Hauskrecht. Conditional Anomaly Detection Using Soft Harmonic Functions: An Application to Clinical Alerting. The 28th International Conference on Machine Learning Workshop on Machine Learning for Global Challenges, Jun 2011, Seattle, United States. ⟨hal-00642313⟩
229 View
141 Download

Share

Gmail Facebook Twitter LinkedIn More