Spatial-temporal Patterns Visual Analysis Based On Air Quality Data

In recent years,air pollution is becoming more and more serious all over the world.Visual analysis is an effective means of analyzing big data,which can help users intuitively explore the inherent laws of data.At present,researchers have proposed various analytical methods to study air quality pollution.However,researches on air quality are mainly focused on the time-varying characteristics of single pollutants,and seldom researchers consider the correlation analysis among pollutants,so as to explore the influence of dimensional subspace on the time-varying pattern distribution.At the same time,the visual analysis methods of air quality data,which ignored the spatial correlation between cities.It is difficult to explore the neighborhood information and temporal sequence of the air quality spatiotemporal patterns.Visual analysis of air quality spatiotemporal patterns,which can helps researchers explore air quality patterns,and provide a scientific basis for air pollution control policies.With the development of science and technology,people can obtain large-scale air quality data by sensors and monitoring stations,which provides a reliable research basis for researchers to analyze air pollution problems.However,It is a great challenge to air quality data mining and analysis: how to extract the implicit spatio-temporal patterns from massive and high-dimensional spatio-temporal data and extract valuable information.In this paper,methods of concentration and continuity are proposed to recognize air quality spatial-temporal patterns,and a visualization technology is introduced to help users explore the spatial-temporal patterns and multi-dimensional characteristics from multiple perspectives.The main contributions of this paper are as follows: Air quality data dimension division based on information entropy and mutual information.Air quality time-varying pattern recognization for a single city based on concentration and continuity.Method of air quality spatial-temporal patterns recognization for urban agglomeration based on mathod of concentration and continuity.We design an interactive visual system to explore spatial-temporal patterns of air quality data.For a single city,feature views are designed to visualize its time-varying patterns. spatial-temporal feature views are designed to interactively explore the spatial-temporal patterns of air quality in geographically adjacent urban agglomerations.