MDGCN: Multiple Graph Convolutional Network Based on the Differential Calculation for Passenger Flow Forecasting in Urban Rail Transit

Algorithm 1

MDGCN.

	Input: Historical passenger flow (described in Section 4.2.);
, (described in Section 3.1.1).
Output: Passenger flow of .
Process:
(1)	for i = 1, 2, …, n do
(2)	by equation (7)
(3)	by equation (8)
(4)	by equations (10) and (11)
(5)
(6)
(7)	end
(8)	(Forecasted passenger flow of all stations)