Time Series Classification
Time series classification focuses on automatically assigning labels to sequential data, aiming to improve accuracy and efficiency across diverse applications. Current research emphasizes developing robust and efficient models, including those based on random forests, convolutional neural networks (CNNs), transformers, and state-space models, often incorporating techniques like transfer learning, multi-objective optimization, and self-supervised learning to address challenges such as limited data, high dimensionality, and noise. These advancements are significant for various fields, enabling improved accuracy in applications ranging from healthcare diagnostics and financial forecasting to environmental monitoring and industrial process control.
Papers
Automatic Feature Engineering for Time Series Classification: Evaluation and Discussion
Aurélien Renault, Alexis Bondu, Vincent Lemaire, Dominique Gay
Enhancing Representation Learning for Periodic Time Series with Floss: A Frequency Domain Regularization Approach
Chunwei Yang, Xiaoxu Chen, Lijun Sun, Hongyu Yang, Yuankai Wu
QUANT: A Minimalist Interval Method for Time Series Classification
Angus Dempster, Daniel F. Schmidt, Geoffrey I. Webb