Short-Term Passenger Flow Forecast of Rail Transit Station Based on MIC Feature Selection and ST-LightGBM considering Transfer Passenger Flow

Table 2

Theoretical analysis and comparison of methods.

Studies	Temporal feature	Spatial feature	Transfer feature	Model	Applicable data size of model	Time complexity	Modeling difficulty	Parameter complexity	Accuracy
Feng and Cai [40]	Yes	No	No	ARIMA	Small-medium	Low	Low	Low	Low
Sun et al. [21]	Yes	No	No	Wavelet SVM	Small	Medium	Medium	Medium	Medium
Jin et al. [41]	Yes	No	No	BP network	Big	High	Medium	High	High
Tang et al. [32]	Yes	Yes	No	ST-LSTM	Big	High	High	High	High
Zhang et al. [42]	Yes	Yes	No	Multi-LSTM	Big	High	High	High	High
The proposed method	Yes	Yes	Yes	ST-LightGBM	Big	Low	Medium	Low	High