Corpus ID: 210713940

Generalization of Change-Point Detection in Time Series Data Based on Direct Density Ratio Estimation

@article{Hushchyn2020GeneralizationOC,
  title={Generalization of Change-Point Detection in Time Series Data Based on Direct Density Ratio Estimation},
  author={M. Hushchyn and A. Ustyuzhanin},
  journal={ArXiv},
  year={2020},
  volume={abs/2001.06386}
}
  • M. Hushchyn, A. Ustyuzhanin
  • Published 2020
  • Computer Science, Mathematics
  • ArXiv
  • 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… CONTINUE READING
    1 Citations

    Figures, Tables, and Topics from this paper.

    Online Neural Networks for Change-Point Detection

    References

    SHOWING 1-10 OF 17 REFERENCES
    Sequential change-point detection based on direct density-ratio estimation
    • 127
    • PDF
    Change-point detection in time-series data by relative density-ratio estimation
    • 135
    • Highly Influential
    • PDF
    A survey of methods for time series change point detection
    • 300
    • PDF
    Real-Time Change Point Detection with Application to Smart Home Time Series Data
    • 27
    • PDF
    Unsupervised Change Analysis Using Supervised Learning
    • 52
    • PDF
    A Least-squares Approach to Direct Importance Estimation
    • 338
    • PDF
    Machine Learning in Non-Stationary Environments - Introduction to Covariate Shift Adaptation
    • 237