Event Detection in Time Series: Universal Deep Learning Approach
This addresses the problem of detecting rare and time-interval events in imbalanced time series data for applications like anomaly detection or monitoring.
The paper tackles event detection in time series by proposing a supervised regression-based deep learning approach to handle imbalanced datasets and rare events, demonstrating superior performance across diverse domains.
Event detection in time series is a challenging task due to the prevalence of imbalanced datasets, rare events, and time interval-defined events. Traditional supervised deep learning methods primarily employ binary classification, where each time step is assigned a binary label indicating the presence or absence of an event. However, these methods struggle to handle these specific scenarios effectively. To address these limitations, we propose a novel supervised regression-based deep learning approach that offers several advantages over classification-based methods. Our approach, with a limited number of parameters, can effectively handle various types of events within a unified framework, including rare events and imbalanced datasets. We provide theoretical justifications for its universality and precision and demonstrate its superior performance across diverse domains, particularly for rare events and imbalanced datasets.