基于加权深度森林的长白山风景区客流量预测

Tourist flow prediction of Changbai Mountain Scenic Area based on weighted deep forest

提出一种改进的加权深度森林方法。主要通过筛选多粒度扫描产生的子样本,使预测结果性能更好。选取长白山风景区客流量数据集作为样本,将随机森林、支持向量机和原始深度森林等作为对比方法,验证了模型的有效性。结果表明,提出的算法R^(2)为0.9586,明显高于其他方法。该算法能有效提高客流量预测的准确性,对旅游业的发展具有重要的理论意义与实用价值。

An improved weighted deep forest method is proposed.Mainly by screening sub-samples generated from multi-grained scanning,the performance of prediction results is made better.The tourist flow dataset of Changbai Mountain Scenic Area is selected as the sample,and random forest,support vector machine and original deep forest are used as the comparison methods to verify the effectiveness of the model,and the results show that the R^(2) of the algorithm proposed in this study is 0.9586,which is significantly higher than other methods.The algorithm can effectively improve the accuracy of tourist flow prediction,which has important theoretical significance of the tourism industry and practical value.