Generalization of Change-Point Detection in Time Series Data Based on Direct Density Ratio Estimation
This work provides an incremental improvement for researchers and practitioners in time series analysis by enhancing change-point detection methods.
The authors tackled the problem of change-point detection in time series data by generalizing existing algorithms based on direct density ratio estimation, using methods like Gradient Boosting over Decision Trees and Neural Networks, and showed that these outperform the classical RuLSIF algorithm on synthetic and real-world datasets.
The goal of the change-point detection is to discover changes of time series distribution. One of the state of the art approaches of the change-point detection are based on direct density ratio estimation. In this work we show how existing algorithms can be generalized using various binary classification and regression models. In particular, we show that the Gradient Boosting over Decision Trees and Neural Networks can be used for this purpose. The algorithms are tested on several synthetic and real-world datasets. The results show that the proposed methods outperform classical RuLSIF algorithm. Discussion of cases where the proposed algorithms have advantages over existing methods are also provided.