摘要
国家公园网络关注度是评估旅游地的发展状况,感知游客活动行为偏好,预测潜在游客流量需求的重要指标。但从长时间、动态时间视角探讨网络关注度周期性与突变性特征方面的研究较少。本研究基于2014—2023年武夷山国家公园百度指数数据,采用季节性强度指数、M-K突变检验、小波分析等多种方法探究网络关注度多维时间变化特征,并利用SARIMA-SVM组合模型预测武夷山国家公园网络关注度未来趋势数值。研究表明:1)武夷山国家公园网络关注度年度变化呈单峰趋势,月度变化呈双峰季节性,4、7月为高峰,12月为低谷。2)武夷山国家公园网络关注度的突变点集中在春节期间和年末,整体呈现长期的上升趋势,四季各周关注度的周期性变化显著。3)网络关注度的时间特征与节假日制度、节事活动、接待旅游人数呈显著正相关关系。4)SARIMA-SVM组合模型在武夷山国家公园网络关注度预测方面具有良好适用性,在关注度突然大幅度上升或下降时间段拟合效果最佳,准确度约为94.92%。本研究旨在明晰国家公园网络关注度时间序列特征并进行未来预测,为应对国家公园现实游客冲击力提供科学依据。
The network attention to national parks is an important indicator for assessing the development status of tourist destinations,perceiving the behavioral preferences of tourist activities,and predicting the demand of potential tourist flows.However,there are fewer studies on exploring the periodic and sudden change characteristics of the network attention from a long time dynamic time perspective.Based on the Baidu index data of the Wuyi Mountain National Park from 2014 to 2023,this study explored the multidimensional time change characteristics of the network attention by using various methods,such as seasonal intensity index,M-K change test,wavelet analysis,etc.,and predicted the future trend values of the network attention to the Wuyi Mountain National Park by using the SARIMA-SVM combination model.The study showed that:1)The annual change of the network attention to the park presented a single peak trend,and the monthly change exhibited a bimodal seasonality,with peaks in April and July,and a trough in December.2)The change points of the network attention to the park concentrated in the period of Spring Festival and at the end of a year with the overall long-term upward trend.The weekly network attention in four seasons presented a significantly periodic change.3)The temporal characteristics of network attention were related to the holiday system,festival activities,and the number of tourists.4)The SARIMA-SVM combination model had good applicability in the prediction of the network attention to the park,with the best fitting effect in the time period of a sudden and substantial increase or decrease in attention,and the accuracy was about 94.92%.This study aims to clarify the time-series characteristics of national park network attention and make future predictions to provide a scientific basis for coping with the real tourist impact of national parks.
作者
张丽甜
崔亭亭
庄迎鑫
朱里莹
ZHANG Litian;CUI Tingting;ZHUANG Yingxin;ZHU Liying(College of Landscape Architecture and Arts,Fujian Agriculture and Forestry University,Fuzhou 350100,Fujian,China)
出处
《西北林学院学报》
北大核心
2025年第2期259-272,共14页
Journal of Northwest Forestry University
基金
国家自然科学基金(32401461)
福建省891号科技专项(115-KLY23110XA)
福建农林大学专项(XJQ2021S2)。
关键词
网络关注度
百度指数
国家公园
时间特征
未来预测
network attention
Baidu index
national park
temporal characterization
future prediction
