Health and air quality data correlation filtering method based on multi-level mutual information

• Main Authors: QIANG XING, PAN WEIDONG, HUA YUEMING, YUE WEIQING
• Format: Patent
• Language: Chinese; English
• Published: 18.10.2019
• Subjects: CALCULATING; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention discloses a health and air quality data correlation filtering method based on multi-level mutual information. The method includes: inputting the human body health data and the air quality index data into an adaptive multi-level mutual information calculation method; scanning and sampling the data of the time sequence corresponding to the input data group layer by layer on the granularity of the multi-level time window through the time window; and calculating a mutual information value of the window through a KSG estimation algorithm, and setting a mutual information threshold value for filtering by using two standardization methods to obtain a window set with correlation between the human body health data and the air quality index data. Compared with a traditional mutual information calculation method, the method is higher in processing accuracy and less limited by input data, and the time for calculating the mutual information under the same hardware condition is shorterthan that of most method