This paper presents a method which uses an artificial neural network (ANN) to forecast international tourism demand to Vietnam. In our work, the monthly statistical data between 2005 and 2016 from Vietnam National Administration of Tourism were used for training, and the data in 2017 was used for testing. The ANN model used in the study is a nonlinear autoregressive neural network where forecasting values are inferred from the previous observed data. Moreover, the ANN is investigated into variety of data preprocessing techniques including deseasoning and detrending. To evaluate the accuracies, the mean of absolute deviation was considered. The experimental results have shown that the use of the ANN with deseasioning gives superior results than the other preprocessing data techniques.