Conditional Anomaly Detection with Soft Harmonic Functions

Michal Valko 1 Branislav Kveton 2 Hamed Valizadegan 3 Gregory Cooper 3 Milos Hauskrecht 3
1 SEQUEL - Sequential Learning
LIFL - Laboratoire d'Informatique Fondamentale de Lille, Inria Lille - Nord Europe, LAGIS - Laboratoire d'Automatique, Génie Informatique et Signal
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.
Complete list of metadatas

Cited literature [27 references]  Display  Hide  Download

https://hal.inria.fr/hal-00641081
Contributor : Michal Valko <>
Submitted on : Tuesday, November 15, 2011 - 3:40:31 PM
Last modification on : Thursday, February 21, 2019 - 10:52:49 AM
Long-term archiving on : Thursday, February 16, 2012 - 2:20:58 AM

File

valko2011conditionala.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00641081, version 1

Collections

Citation

Michal Valko, Branislav Kveton, Hamed Valizadegan, Gregory Cooper, Milos Hauskrecht. Conditional Anomaly Detection with Soft Harmonic Functions. Proceedings of the 2011 IEEE International Conference on Data Mining, Dec 2011, Vancouver, Canada. ⟨hal-00641081⟩

Share

Metrics

Record views

437

Files downloads

245