Tourism demand time sequence prediction method based on guiding attention mechanism

Abstract

The invention relates to a travel demand time sequence prediction method based on an attention guiding mechanism, and belongs to the field of travel demand prediction. The method comprises the following steps: S1, combining travel demand time sequences from a plurality of countries and regions; S2, defining a partial order relationship to compare the stability among different time sequences, defining homogeneous polynomials of different dimensions to combine the time sequences, and then adjusting the stability of the travel demand time sequences by using a guide attention mechanism; S3, constructing a full-connection long and short memory neural network, and guiding an attention mechanism to allow the network to perform parameter training on different training data according to needs; S4, predicting a travel demand time sequence by using the trained full-connection long-short memory neural network as a prediction model, and dividing the travel demand time sequence according to a partial order relationship; and S5, solving the prediction model, and carrying out back calculation to obtain the tourism demand of each country or region. The performance of the prediction model is improved.

Description

technical field [0001] The invention belongs to the field of tourism demand prediction, and relates to a time series prediction method of tourism demand based on a guiding attention mechanism. Background technique [0002] In recent years, various tourism demand forecasting models have been proposed to forecast tourism demand. Existing popular models are divided into two categories according to their structures, namely time series analysis models and machine learning models. The establishment of the above two models is to learn the characteristics of time series, and can compare the effect of tourism demand forecasting in different regions. [0003] Traditional machine learning models for tourism demand forecasting, compared with time series analysis models, have a certain age based on machine learning methods because they can capture nonlinear relationships between data sets, especially in terms of long-term forecasting, such as neural networks Optimal Forecasting Using A...

AI Technical Summary

Problems solved by technology

First of all, artificial prior information is difficult to quantify, and tourism demand forecasting requires high-precision parameter adjustment

Images