在过去，克拉科夫这座古城被称为波兰的首都。2000年，它被称为官方的欧洲文化之都。现在，它因拥有欧洲最污染的空气而闻名......在世界卫生组织（WHO）的一项研究中，克拉科夫被评为世界上污染最严重的城市之一。在该报告中，该市的PM2.5水平在575个城市中排名第8，PM10水平在1100个城市中排名第145。危险的空气质量是一个常见的问题，特别是在较冷的月份，许多居民使用固体燃料（主要是煤）进行家庭取暖。克拉科夫的空气污染对人类健康和生命构成了巨大的威胁。克拉科夫的有毒空气包括：颗粒物、苯并(a)芘和二氧化氮。

国家管理的监测站网络由克拉科夫的8个监测站组成。我们决定更进一步--建立低成本的空气质量传感器网络，可以部署在整个城市。与雾霾作斗争的第一步是确定问题区域，提高居民和当局的意识。建立一个传感器网络是非常重要的--只有这样你才能检查城市各个地区的实际情况。该技术能够通过map.airly.eu对空气质量进行实时监测，因此关于特定地点的空气的信息很容易获得，而且总是最新的。

Content

这个数据集包括2017年的空气质量数据（颗粒物PM1、PM2.5和PM10的浓度、温度、气压和湿度），由位于波兰克拉科夫的56个低成本传感器网络产生。

每个人都有自己的位置（其中6个在这段时间内被替换，几乎有相同的纬度和经度）。

测量结果被归纳为12个文件，每个月一个。数据的分辨率为1小时。

Known issues:

• PM1 is not calibrated and therefore can be bigger than PM2.5

• PM2.5 can be bigger than PM10 within the limits of measurement error

• for the first two months humidity and temperature were not calibrated and therefore can show inaccurate values

Acknowledgements

The data was generated by Airly network - the project is still in its beginning stage, but over 1000 sensors have already been implemented in Poland. Airly is a startup definitely worth watching, especially for citizens of the most polluted cities. After all, it’s clean air we all want to breathe.

Inspiration

I think that this dataset offers some great opportunities for predictive models and data visualization. Airly's goal is to develop an effective forecast and monitoring of air quality, employ Artificial Intelligence and utilise data from extensive sensor network. If anyone has any ideas, breakthroughs or other interesting models please post them.

Some questions worth exploring:

• What are the best prediction models based on extensive sensor network - statistical or numerical forecast?

• How weather affects air quality?

• How much pollution comes from cars, factories and coal-fired power plants?