A Combination Model Based Deep Long Term Model for Tourism Demand Forecasting.

The accurate tourism demand forecasting is crucial to the development of tourism. However, the challenge of non-linear features recognizing in tourism time series makes it a troublesome thing. To overcome the above difficulties, this paper proposes a novel model for tourism demand forecasting based on a long term recurrent neural network with an evolutionary optimization algorithm. The model aims to learn the features of tourism demand time series by combining several sequences, and the proposed model consists of two sections, the first section defines the employed neural network; the second section introduces the optimization algorithm to search the optimal weights for difference sequences. Tourism demand time series of Macau has been adopted to validate the proposed model, and the experimental results show that the proposed method can accurately forecast the daily tourism demand of Macau, China.