Abstract: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.

What problem does this paper attempt to address?

This paper attempts to address the challenges of detecting events in time - series data, especially for imbalanced data sets, rare events, and events defined by time intervals. Traditional supervised deep - learning methods mainly adopt a binary - classification framework, that is, each time point is assigned a binary label indicating the presence or absence of an event. However, these methods are not effective in handling the above - mentioned specific scenarios. To overcome these limitations, the author proposes a new regression - based supervised deep - learning method, which has the following advantages: 1. **Handling various types of events**: The new method can effectively handle different types of events, including rare events and imbalanced data sets. 2. **Unified framework**: This method processes various events within a unified framework, reducing the number of parameters and improving the efficiency of the model. 3. **Theoretical basis**: The author provides the theoretical basis of this method, proving its universality and accuracy. 4. **Practical performance**: Through tests on actual data sets in multiple fields, this method has demonstrated its superior performance, especially in the detection of rare events and imbalanced data sets. Specifically, the main contributions of this paper are: - Proposing a regression - based deep - learning method for event detection in time series. - Proving the universality and accuracy of this method through theoretical analysis. - Verifying the effectiveness of this method on multiple actual data sets, especially performing well in tasks such as fraud detection and bow - shock - crossing identification. This method not only has a solid theoretical foundation but also shows strong performance in practical applications, especially in handling imbalanced data sets and rare events. This makes this method an important progress in the field of time - series event detection.

Event Detection in Time Series: Universal Deep Learning Approach

A Comprehensive Python Library for Deep Learning-Based Event Detection in Multivariate Time Series Data and Information Retrieval in NLP

Deep Learning for Time-Series Analysis

An Enhancing Timeseries Anomaly Detection Using LSTM and Bi-LSTM Architectures

Deep Learning for Time Series Anomaly Detection: A Survey

Deep-Eware: spatio-temporal social event detection using a hybrid learning model

Unsupervised Deep Learning for IoT Time Series

Abnormal Events Detection Using Deep Networks for Video Surveillance.

A Survey of Deep Learning and Foundation Models for Time Series Forecasting

Modeling Extreme Events in Time Series Prediction

Deep learning for time series classification: a review

Multivariate Time Series Classification: A Deep Learning Approach

Soccer Event Detection Using Deep Learning

Abnormal event detection via multi-instance dictionary learning

Generalized durative event detection on social media

Deep Learning for Time Series Forecasting: A Survey

Unsupervised Event Outlier Detection in Continuous Time

Deep Learning for Time Series Classification and Extrinsic Regression: A Current Survey