02-06-2023 дата публикации
Номер: CN116205329A
Принадлежит:
The invention discloses a holiday and festival passenger flow volume prediction method. The method comprises the following steps of (1) performing statistics on holiday and festival passenger flow volume historical data in recent three years; (2) cleaning the historical data; (3) constructing time sequence models of an aperiodic change trend term, a seasonal trend term, a holiday trend term, a distance trend term, an epidemic situation trend term, an error term and a weight term occupied by each trend term based on a prophet algorithm; (4) substituting the cleaned historical data into each time sequence model for training; and (5) carrying out passenger flow volume prediction based on the trained model. According to the method, optimization is carried out on the basis of a traditional prophet algorithm, an epidemic situation factor and a distance factor are added, meanwhile, different weights are set for different trend terms, an optimized prophet + + model is obtained, the real situation ...
Подробнее