基于大数据及网络信息提取对空气质量进行分析及预测The analyze and forecast to air quality based on big data and network information extract 内容摘要□□ 在当前我国城市化进程的快速推进下，我国的交通规模、能源消耗也在不断扩大，一氧化碳等有毒气体及固体污染物大量增加，严重影响了人们的正常生活，如何减少空气污染、打好污染防治攻坚战，对推动生态文明建设有很强的指导性。所以我从国内某的空气质量记录网站上记录的 384 个城市通过网络信息爬取获得了从 2014 年至今的空气质量记录数据，包括一氧化碳浓度、二氧化硫浓度等参数，然后对数据中的缺失值、异常值进行处理，数据可视化、分析数据特征、接着对每个城市先按照省份进行分组，并将数据保存到数据库中，同时对每个城市采用时间序列分析，在使时序数据变得稳定后，对时序数据进行预测，用户可以通过输入城市和起止日期来预测这段时间的空气质量指数。□关键词：□空气质量预测□ARIMA 模型□网络信息爬取□时序分析Abstract□With the rapid promotion of our countries urbanization rate、 the urban traffic scale and energy consumption are also enlarging rapidly、 which raise a plenty of toxic gas and solid grain contamination、 such as sulfur dioxide and fine particulate matter 、 respirable solid pollutants and carbon monoxide. These pollutant make a serious influence to humans’ normal life. So 、 reduce the air pollution and winning the Pollute prevention and management battle would have a great instructive for promoting the Ecological Civilization Construction. Therefore、 I got three hundred and eighty four cities air quality index data 、 including carbon monoxide concentration 、 sulfur dioxide concentration and so on since 2014 to now through the Network information crawling technology recorded on the air quality records monitor on-line website. Then processing the air quality index data by dealing the missing and abnormal values. After that、 dealing the data by making data visualization and a...
