Cleaning method of health data for atmospheric pollution health risk assessment

Abstract

The present invention provides a cleaning method of health data for atmospheric pollution health risk assessment. The cleaning method comprises five steps: 1, extracting a few samples in health data, determining whether patients are local permanent residents or people who temporarily come here and seek treatment through manpower according to patients' registered residences and present address information registered in the samples; 2, designing six rules, and constructing a decision tree; 3, processing a data set of the samples based on the six rules, and obtaining the training data set of the decision tree through the combination of the manual mark results in the step 1; 4 constructing the decision tree according to the training data set; and 5, performing the determination of the six rules in the steps to the cleaning data in turn, inputting six results into the decision tree trained in the step 4, and obtaining a final determination result. The cleaning method of health data for atmospheric pollution health risk assessment has scientific conception, simple calculation and high universality, has good practical values in the health data cleaning for atmospheric pollution health risk assessment, and has practical application prospects in the popularization of the real work of public health and environmental health.

Description

technical field [0001] The invention relates to a health data cleaning method for air pollution health risk assessment. The method cleans the health data from different medical institutions in the same city, and extracts the health data with the highest probability from the health data with different structures and vague descriptions. A specific health dataset of the city's permanent population is used to analyze and evaluate the relationship between air pollution and health risks in the city. It belongs to the technical field of public health and environmental health. Background technique [0002] Due to the large geographic differences in meteorological characteristics and major air pollutants, the health impact of air pollution in a specific city or region is more concentrated on the permanent population in that region. Therefore, the analysis and assessment of air pollution health risks in a specific city or region needs to be based on the health data of the resident po...

AI Technical Summary

Problems solved by technology

[0006] In order to solve the problem of fuzzy and inaccurate information about the registered residence and current address of patients in the medical data of patients from various medical institutions in the same area used in air pollution health risk assessment, an automatic health data cleaning based on decision tree learning is proposed method, which is used to distinguish whether the patients are local permanent residents or temporary local medical patients

Images