Holiday Travel Pattern Forecast Based On Machine Learning Algorithm

摘要

This paper uses multiple machine learning algorithms, such as random forest, SUV, Logistic, decision tree, BP neural network etc. to predict the traffic conditions and environmental conditions for travel pattern forecast, then compare the results to obtain the optimal algorithm for classification prediction. At the same time, a visitor flow forecasting model is built based on the wavelet analysis and ARIMA algorithm to predicts and analyzes the visitor traffic in the tourist scenic spot.