摘要
探索数据源"百度指数"下艾滋病疫情的时空分布及防控要点.应用爬虫技术编写JAVA程序来挖掘数据库数据,引进向量自回归模型原理,分析了艾滋病发病数与各个主成分的Granger因果关系.应用面板数据模型,对艾滋病疫情进行了时间及空间定性、直观的分析.结果表明官方发布的艾滋病发病数与艾滋病疑似患者的搜索行为呈滞后的联动关系,且不同地区、不同时间的艾滋病发病数存在显著差异.
The aim is to explore the spatial-temporal distribution of the AIDS epidemic with the data source of Baidu-index and give the key points of monitor. By applying the crawler technology to write a JAVA program to mining data and making use of the basic idea of Vector Auto-regression to analyze the relationship between the three main components and AIDS. Using panel data model to explore the AIDS epidemic, and give a qualitative an intuitive analysis. The result points that the official release about the AIDS incidence is significantly lagged effect with the AIDS-related terms in Baidu index, and different areas and time show significant differences.
出处
《福建师范大学学报(自然科学版)》
CAS
CSCD
北大核心
2016年第3期14-18,共5页
Journal of Fujian Normal University:Natural Science Edition
基金
福州市社会科学研究规划课题资助项目(2015C03)
关键词
百度指数
时空分析
面板数据
艾滋病
Baidu-index
spatial-temporal analysis
panel data
AIDS