BP神经网络的样本拓展及旅游人数预测

Sample Expanding of BP Neural Network and Forecasting Number of Tourists

城市旅游人数预测是一个城市建设规划中的重要决策因素,神经网络恰能描述其非线性的特点。为了使算法在对不同城市的预测上具有通用性,只根据往年旅游人数这一单一指标进行预测,有效地避开了其他因素对其的影响。同时,通过对网络的样本拓展,增加了预测数据的有效性。实验结果表明,该预测模型可以解决城市旅游人数的预测问题,验证了模型的可行性和通用性。

Forecasting number of tourists is an important factor in decision - making of urban construction. This forecasting is a nonlinear question, just can be described by BP neural network. In order to algorithm proposed by this paper can apply to different cites＇ forecasting, BP neural network only choose one single factor, namely the number of tourists about past year. This avoid the influence of other factor effectively. Meanwhile, availability of forecasting data be increased through the sample expanding. The experiment results show that the model is well - suited for forecasting the number of tourists, and the feasibility and effectiveness of its is verified.