Design issues in Time Series dataset balancing algorithms

TSVNet: Combining Time-Series and Opportunistic Sensing by Transfer Learning for Dynamic Thermal Sensation Estimation

“…However, we cannot apply SMOGN to our regression problem with the time-series input because SMOGN does not support the generation of time-series data. Therefore, we conduct data balancing by combining SMOTER with TS_SMOTE [48], which is a time series data generation method for two-class classification.…”