Research On Tourism Big Data Enabling Tourist Travel Prediction Model

The analysis of the timeliness and accuracy of tourist flow is an important research problem in the construction of smart tourism.The analysis of tourist flow is helpful to the relevant departments of tourism management to effectively manage tourism activities,as well as to the construction and development of local catering,accommodation,travel,play,shopping,entertainment and other aspects.At present,the tourism industry has gradually developed into the pillar industry of our country’s national economy,while Dali is a typical tourist city;tourism is Dali’s pillar industry and its GDP proportion is relatively large.The study of tourist flow in Dali can improve and strengthen the construction of Dali tourism,which is of great significance to the economic development of Dali.Big data generated based on the Internet and mobile Internet has become an important data source for tourism research.Tourism big data generated on various Internet platforms has a fast generation speed,high data frequency,multiple data types,and contains rich and valuable information.The purpose of this paper is to build a Dali tourist flow prediction model based on the Baidu search index of Dali tourism keywords in the tourism big data,and improve the accuracy of the model prediction through research and analysis.There are three innovations in this paper: first,it uses monthly data to study Dali tourist flow;second,it uses text mining method to expand Dali tourism keyword database;third,it uses the idea of comparing various models to improve the accuracy of prediction.The main research methods of this paper are as follows:(1)Using Python and Octopus data crawling software to analyze the text content of travel notes and comments in the Internet tourism platforms and software such as Ctrip and Hornet’s Nest,analyze the text data about Dali tourism,construct the relevant keyword database,and obtain the daily degree data of the keyword search index of Baidu;(2)Pearson correlation coefficient was used to analyze the keyword database to select the keywords suitable for Dali tourist flow prediction;(3)Build a Dali tourist flow prediction model based on Baidu search index for analysis,including VAR model and machine learning model,and combine the advantages of each model to build a combined model for more accurate prediction.This paper draws the following conclusions: the Baidu search index of Dali tourism keywords in tourism big data can predict the change of Dali tourist flow to a certain extent;Compared with the single model,the combined model using machine learning algorithm can fully extract the change information of tourist flow in peacetime and holidays,and the prediction accuracy is greatly improved.Compared with the three models,it is found that the MLP neural network has the highest accuracy.