Table 1 An overview of the forecast models
Model | Describe | Strengths |
|---|---|---|
AdaB | an ensemble learning algorithm by iteration until a stop condition is reached or the error rate becomes sufficiently small [27]. | the ability to handle complex datasets and feature interactions |
LGB | based on gradient boosting decision trees | optimize training speed and memory usage |
XGB | a boosting integrated machine learning algorithm based on the CART regression tree. | integrates regularization techniques and feature selection methods, demonstrating strong generalization ability and predictive performance [17]. |
CatB | a gradient boosting machine learning algorithm | high performance in categorical features |
LR | a supervised learning method and a member of the general linear model family [16] | simple |
LSTM | a supervised recurrent neural network | capture time correlation more effectively [16]. |
MLP | one of the simplest artificial neural networks (ANNs) for data classification tasks [17] [17]. | suitable for solving classification and regression problems |