Anomaly Detection Based on Aggregation of Indicators

TL;DR

This paper presents a methodology for anomaly detection that combines expert knowledge and feature selection to classify monitoring signals, aiding human operators in identifying the origin of anomalies.

Contribution

It introduces a general approach leveraging large indicator sets and feature selection with Naive Bayes classification for anomaly detection and origin identification.

Findings

Effective in simulated engine anomaly data

Improves anomaly classification accuracy

Automates indicator selection process

Abstract

Automatic anomaly detection is a major issue in various areas. Beyond mere detection, the identification of the origin of the problem that produced the anomaly is also essential. This paper introduces a general methodology that can assist human operators who aim at classifying monitoring signals. The main idea is to leverage expert knowledge by generating a very large number of indicators. A feature selection method is used to keep only the most discriminant indicators which are used as inputs of a Naive Bayes classifier. The parameters of the classifier have been optimized indirectly by the selection process. Simulated data designed to reproduce some of the anomaly types observed in real world engines.