Training LSTMS with circular-shift epochs for accurate event forecasting in imbalanced time series.
Xiaoqian ChenLalit GuptaPublished in: Expert Syst. Appl. (2024)
Keyphrases
- weather forecasting
- financial time series
- forecasting accuracy
- short term
- long term
- high quality
- exponential smoothing
- training set
- arma model
- non stationary
- high accuracy
- hough transform
- event detection
- stock market
- support vector regression
- dynamic time warping
- temporal patterns
- news articles
- supervised learning
- event types
- hidden markov models
- test set
- arima model
- chaotic time series
- demand forecasting
- computationally efficient
- training samples
- forecasting model