Predictive change point detection for heterogeneous data

TL;DR

This paper introduces a predictive change point detection framework using machine learning models like ARIMA and LSTM, outperforming traditional methods in false positive rate and detection accuracy, demonstrated through a tribological case study.

Contribution

The paper presents a novel predictive change point detection framework that improves detection quality by integrating advanced predictive models and comparing their forecasts with actual data.

Findings

Outperforms state-of-the-art online CPD routines in false positive rate

Effective in detecting change points with few false positives

Demonstrated success in a tribological wear case study

Abstract

A change point detection (CPD) framework assisted by a predictive machine learning model called "Predict and Compare" is introduced and characterised in relation to other state-of-the-art online CPD routines which it outperforms in terms of false positive rate and out-of-control average run length. The method's focus is on improving standard methods from sequential analysis such as the CUSUM rule in terms of these quality measures. This is achieved by replacing typically used trend estimation functionals such as the running mean with more sophisticated predictive models (Predict step), and comparing their prognosis with actual data (Compare step). The two models used in the Predict step are the ARIMA model and the LSTM recursive neural network. However, the framework is formulated in general terms, so as to allow the use of other prediction or comparison methods than those tested here. The power of the method is demonstrated in a tribological case study in which change points separating the run-in, steady-state, and divergent wear phases are detected in the regime of very few false positives.