Method and system for predicting passenger flow volume of scenic spot, and storage medium

Abstract

The embodiments of the invention provide a method and a system for predicting the passenger flow volume of a scenic spot, and a storage medium, belonging to the technical field of data prediction. Themethod comprises the steps of: acquiring multi-column data of the scenic spot; defining an attention weight population; weighting multi-column data; inputting weighted multi-column data into a GRU neural network so as to obtain a corresponding prediction value; respectively calculating an error between each predicted value and a corresponding standard value; screening out the first two multi-column data with a minimum error from the plurality of weighted multi-column data; randomly selecting a weight value from the attention weights of the two screened multi-column data respectively; carryingout gene recombination; carrying out gene mutation; updating the attention weight population; updating the number of iterations; judging whether the number of iterations is greater than or equal to apreset threshold value or not; outputting the attention weight with a minimum error as an optimal solution; and adding the optimal solution into a GRU model so as to predict the passenger flow volumeof the scenic spot. The method can provide prediction precision of the passenger flow volume of the scenic spot.

Description

technical field [0001] The present invention relates to the technical field of data prediction, in particular to a method, system and storage medium for predicting passenger flow in a scenic spot. Background technique [0002] Short-term passenger flow forecasting in scenic spots is one of the key issues in current scenic spot management. The current short-term passenger flow forecasting models for scenic spots include traditional time series models, BP neural network models, support vector machines, etc., and their models after algorithm optimization. Short-term passenger flow forecasting has very strong nonlinear characteristics, and the current model's ability to deal with strong nonlinear characteristics is far from enough. Moreover, passenger flow forecasting is affected by many factors, and the importance and degree of influence of each factor are different. All current models do not take this issue into account, which may lead to low forecasting accuracy. Contents ...

AI Technical Summary

Problems solved by technology

Short-term passenger flow forecasting has very strong nonlinear characteristics, and the current model's ability to deal with strong nonlinear characteristics is far from enough

Moreover, passenger flow forecasting is affected by many factors, and the importance and degree of influence of each factor are different. All current models do not take this issue into account, which may lead to low forecasting accuracy.

Images