What question did this study set out to answer?

The aim is to address imbalanced class distributions in time series classification using a novel oversampling method.

May 6, 2026Applied Sciences0 citationsOpen Access

Synthetic Minority Oversampling for Imbalanced Time Series Classification Based on Path Signature

MAMohnad Abunada SBSamir Brahim Belhaouari HBHalima Bensmail

Key Points

The aim is to address imbalanced class distributions in time series classification using a novel oversampling method.
Introduced Path Signature Synthetic Time-series Oversampling (PSSTO) to synthesize minority samples.
Applied PSSTO in path signature space while pruning low-quality samples.
Evaluated performance across 12 public datasets using random forest classification.
PSSTO with random forest shows significant classification improvement over conventional methods.
Statistical analysis indicates significant gains using pairwise Wilcoxon signed-rank tests.
Achieved best averages in F1, G-mean, and AUC compared to other time series-specific oversamplers.

Abstract

Imbalanced class distributions hinder time series classifiers by underrepresenting rare yet important events. We introduce Path Signature Synthetic Time-series Oversampling (PSSTO), a structure-preserving oversampling method that operates in path signature space to synthesize informative minority samples while pruning low-quality ones. Across 12 public datasets, PSSTO with a random forest improves classification over conventional resampling approaches on average. Pairwise Wilcoxon signed-rank tests against these approaches indicate statistically significant gains. Compared with time series-specific oversamplers, PSSTO with random forest attains the best averages on F1, G-mean, and AUC compared to the strongest alternative. These results show that structure-preserving oversampling in signature space is an effective and broadly applicable remedy for imbalanced time-series classification.

问 AI

Bookmark

View Full Paper