TY - GEN
T1 - Data-driven regional analysis of urban atmosphere pollution based on density clustering
AU - Wu, Xiaoting
AU - Wang, Qinglin
AU - Liu, Yu
AU - Li, Yuan
N1 - Publisher Copyright:
© 2017 IEEE.
PY - 2017/7/2
Y1 - 2017/7/2
N2 - Severe haze or smog episodes have been frequent in China during the recent decades. Air pollution data with both temporal and spatial characteristics can be analyzed by using data mining technology to explore hidden knowledge. For the complex urban air pollution situation in a large scale, a data-driven urban air pollution division method based on clustering and association rules mining is proposed in this study. Firstly, an improved association rule mining method is proposed and used to analyze the relationships between AQI and weather conditions. Then the DCBV (Density clustering based on Voronoi) method is proposed for city point clustering. Lastly, a similarity calculation model of urban air pollution is presented and combined with DCBV to cluster cities according to their air pollution situation. The proposed method has been applied in analyzing the air pollution data which is obtained from one-year monitoring in 348 cities in China. Because the cities are divided into groups driven by air pollution data instead of administrative affiliation, the obtained clusters by this method are consistent with air pollution condition, and can provide reference and support for controlling air pollution by region.
AB - Severe haze or smog episodes have been frequent in China during the recent decades. Air pollution data with both temporal and spatial characteristics can be analyzed by using data mining technology to explore hidden knowledge. For the complex urban air pollution situation in a large scale, a data-driven urban air pollution division method based on clustering and association rules mining is proposed in this study. Firstly, an improved association rule mining method is proposed and used to analyze the relationships between AQI and weather conditions. Then the DCBV (Density clustering based on Voronoi) method is proposed for city point clustering. Lastly, a similarity calculation model of urban air pollution is presented and combined with DCBV to cluster cities according to their air pollution situation. The proposed method has been applied in analyzing the air pollution data which is obtained from one-year monitoring in 348 cities in China. Because the cities are divided into groups driven by air pollution data instead of administrative affiliation, the obtained clusters by this method are consistent with air pollution condition, and can provide reference and support for controlling air pollution by region.
KW - air pollution data mining
KW - association rules
KW - clustering
UR - http://www.scopus.com/inward/record.url?scp=85046680507&partnerID=8YFLogxK
U2 - 10.1109/ITNEC.2017.8284764
DO - 10.1109/ITNEC.2017.8284764
M3 - Conference contribution
AN - SCOPUS:85046680507
T3 - Proceedings of the 2017 IEEE 2nd Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2017
SP - 412
EP - 417
BT - Proceedings of the 2017 IEEE 2nd Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2017
A2 - Xu, Bing
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2nd IEEE Information Technology, Networking, Electronic and Automation Control Conference, ITNEC 2017
Y2 - 15 December 2017 through 17 December 2017
ER -
