Tourism is very important for many countries. Many tourism demand forecasting methodologies are continuously being proposed. Most studies have used lagging economic factors as predictors, but these can cause an inaccurate prediction when unexpected events happen. In this study, a tourism social network will be used in our forecasting model. In addition, a least square support vector regression with genetic algorithm will be developed to predict the monthly tourist arrivals. Grey Relational Analysis indicates that the model outperforms the comparison models, and the null hypothesis of the predicted series having the same mean of the actual series is accepted. The experimental results indicate that the predictors from social network are excellent alternatives to economic indicators.

旅游业对许多国家来说非常重要。不断提出许多旅游需求预测方法。大多数研究使用滞后的经济因素作为预测因素，但是当发生意外事件时，这些因素可能导致预测不准确。在这项研究中，旅游社会网络将用于我们的预测模型。另外，将开发具有遗传算法的最小二乘支持向量回归来预测每月的游客到达量。灰色关联分析表明该模型优于比较模型，并且接受了与实际序列均值相同的预测序列的原假设。实验结果表明，社交网络的预测指标是经济指标的绝佳替代方案。