A Hierarchical Passenger Mobility Prediction Model Applicable to Large Crowding Events

Table 1

Parameter and variable notations.

Parameters	Description
	Trip generation index
	Trip attraction index
()	If the passenger will make the th trip in the day
	The feature set that includes the date information and the passenger’s historical mobility information
()	The departure time, the origin, and the destination of the th trip in the day
	The date features
	The th cluster
Variables	Description
	The joint probability of the first trip for a passenger in the th cluster
	The joint probability of the later trip for a passenger in the th cluster
	The conditional probability that the predicted origin is when the context is () for a passenger in the th cluster
	The time-smoothed counting function of the passenger
	The expectation of prior probability
	The number of times that emerges in the training data of the passenger
	The collection of contexts adjacent to context () of the passenger
	The collective conditional probability that the predicted origin is when the context is () for the passengers in the th cluster
	The collective time-smoothed counting function for the passengers in the th cluster
	The number of times that emerges in the training set for the passengers in the th cluster
	The collection of contexts adjacent to context () for a passenger in the th cluster