A Hybridly Optimized LSTM-Based Data Flow Prediction Model for Dependable Online Ticketing

Table 4

Important model parameters.

Model	Important parameters
LightGBM	Boosting type: GBDT, evaluation metric: RMSE, learning rate: 0.3, min child weight: 3.0, number of iterations: 152
XGBoost	Booster: GBTree, evaluation metric: RMSE, gamma: 0.55, max depth: 19, min child weight: 1.0, number of estimators: 26.0
Random forest	Bootstrap: true, max depth: 5, min sample leaf: 2, min sample split:2, number of estimators: 30
SVR	Kernel: RBF, C: 1, epsilon: 0.1
SARIMA	Autoregressive model (), difference (), moving average (): (2, 0, 24)