SUPPORT VECTOR MACHINE IN FORECASTING TOURISM DEMAND
研究了支持向量回归在旅游需求时间序列预测中的应用，使用2004年4月至2007年1月中国国际游客月度数据建模，并与BP神经网络对比，发现支持向量回归表现更优。
Considering influent factors are difficult to research and the amount of sample is small in forecasting the tourism demand, a new technique——support vector regression, to be applied to forecasting time series of tourism demand, was discussed in the paper. The monthly date set of international tourist arrivals to China during 2004.4-2007.1 was employed to establish the model of SVR. Though comparing with the model of BPNN, the result indicates the SVR was better than BPNN.