基于神经网络算法的海南旅游需求测估模型

Forecast Model for Tour Demand of Hainan Province Based on the Neural Network Algorithm

在详细调查海南旅游相关数据的前提下,先建立模型对海南旅游需求进行了预测,然后分析了影响旅游需求的主要因素.先用GM(1,1)灰色模型对海南省旅游人数进行预测,并用马尔科夫链修正误差,在灰色模型的基础上进行了优化.进一步,我们将灰色模型与BP神经网络模型结合起来进行预测,并针对BP网络输入层提供了2种方法:三年滚动预测、多因素预测.得出结论:海南旅游人数还将会逐年递增.同时,通过比较相对误差发现,对于问题的预测精度:BP神经网络>灰色模型.最后,我们利用灰色关联度模型得出各因素对旅游需求的影响:服务>交通>景观发展>消费>环境.

In this paper,we establish model to forecast the tour demand of Hainan province and analyze the main factors which are affecting the tour demand based on the detailed data about Hainan Tourism.Firstly,we use GM（1,1） Grey Prediction model to forecast the number of tourists of Hainan,and amend the error by the Markov Chain,then optimize the model based on the Grey model.Moreover,we combine the Grey model and the BP Neural Network model.For the BP network input floor,we provided 2 methods：the Three-year Rolling Forecast and the Multiple Factors Forecast.In conclusion,the number of tourists of Hainan can be increased year by year.BP Neural Network model is more accurate than the Grey Prediction model.And the main factors based on the Grey Correlation Degree are（in the order of the importance）service,transportation,development,resume and environment.