【摘要】 针对传统的基于历史数据预测旅游景点人气指数方法在预测精度方面的不足,提出了基于区域差异化的人气指数预测方法.考虑到由于区域差异而导致的各类因素对景点人气指数的影响,需要先对景点之间的区域差异化水平进行测度;在此基础上对旅游景点内部各个影响因素值做预处理,并形成景点人气指数监测数据集合;将合并的景点人气监测值输入极限学习方法模型作为模型输入项,经过中间隐层的反复学习和训练能够得出最优解;鉴于提出方法的中间隐层节点数量已经被事先设定完成,因此具有较强的数据泛化处理能力,也能够得到最优的唯一解.实验数据表明,与传统预测方法相比提出方法的预测值走向趋势更接近于真实值,预测精度更高.更多还原

【Abstract】 Aiming at the shortcoming of predicting the prediction accuracy of the popular index method of tourist attractions based on historical data, the prediction method of popular index based on regional differentiation is put forward. Considering the influence of various factors on the popularity index of attractions due to regional differences, it is necessary to measure the level of regional differentiation between attractions first, and on this basis, the value of various influencing factors within tourist attractions is preprocessed, and the monitoring data set of attraction popularity index is formed. The integrated attraction monitoring value input limit learning method model is used as the model input, and the optimal solution can be obtained through repeated learning and training in the middle implicit layer. In view of the number of intermediate hidden layer nodes proposed in the method, it has strong data generalization processing ability, and can get the optimal unique solution. Experimental data show that the predictive value trend of the proposed method is closer to the real value than the traditional prediction method, and the prediction accuracy is higher.更多还原