Abstract
This article aims to explore a more suitable prediction method for tourism complex environment, to improve the accuracy of tourism prediction results and to explore the development law of China’s domestic tourism so as to better serve the domestic tourism management and tourism decision-making. This study uses grey system theory, BP neural network theory, and the combination model method to model and forecast tourism demand. Firstly, the GM (1, 1) model is established based on the introduction of grey theory. The regular data series are obtained through the transformation of irregular data series, and the prediction model is established. Secondly, in the structure algorithm of the BP neural network, the BP neural network model is established using the data series of travel time and the number of people. Then, combining BP neural network with the grey model, the grey neural network combination model is established to forecast the number of tourists. The prediction accuracy of the model is analyzed by the actual time series data of the number of tourists. Finally, the experimental analysis shows that the combination forecasting makes full use of the information provided by each forecasting model and obtains the combination forecasting model and the best forecasting result so as to improve the forecasting accuracy and reliability.
Highlights
In recent years, tourism has sprung up
It is the first condition to realize the sustainable and healthy development of tourism to establish a scientific and operable tourism demand prediction model and make an accurate and effective prediction. Both at home and abroad, the research of the tourism demand forecasting model method has made considerable achievements, but from the perspective of forecasting accuracy, there is still no universally applicable method [3]. e main reason is that tourism demand itself is a complex system, which is restricted by many factors
According to the current method of combining grey model with the neural network, some scholars put forward three forecasting models: parallel grey neural network combined model, series grey neural network combined model (SGNN), and embedded grey neural network combined model (EGNN) [4, 5]. e so-called combination forecasting of the parallel combination model is to make a weighted average combination of the predicted values obtained by different methods so as to get a more accurate prediction value [1]
Summary
Received 3 March 2021; Revised 20 April 2021; Accepted 28 April 2021; Published 7 May 2021. Is study uses grey system theory, BP neural network theory, and the combination model method to model and forecast tourism demand. In the structure algorithm of the BP neural network, the BP neural network model is established using the data series of travel time and the number of people. En, combining BP neural network with the grey model, the grey neural network combination model is established to forecast the number of tourists. E prediction accuracy of the model is analyzed by the actual time series data of the number of tourists. The experimental analysis shows that the combination forecasting makes full use of the information provided by each forecasting model and obtains the combination forecasting model and the best forecasting result so as to improve the forecasting accuracy and reliability
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
