correlation analysis-based abnormal sign miner discrimination method combining EN with MPA-SVM

Abstract

The invention relates to a correlation analysis-based abnormal sign miner discrimination method combining EN with MPA-SVM. The method comprises the following steps: (1) collecting miner occupational health examination data, and constructing a miner sign parameter data set; (2) randomly dividing the miner sign data into a training set and a prediction set; (3) carrying out normalization processing on data of the training set and the prediction set; (4) adopting a Pearson's correlation coefficient to analyze and delete the physical sign data with high correlation; (5) removing redundant sign information by using EN; and (6) establishing an MPA-SVM miner abnormal sign discrimination model, wherein an evaluation index of prediction set data is used for analysis and evaluation of model performance. according to the method, the correlation analysis EN is combined with the MPA-SVM to be used for recognizing the miners with abnormal signs, the purpose of early-stage accurate screening is achieved for detection of occupational diseases and suspected occupational diseases of the miners. The method is suitable for the field of occupational health auxiliary diagnosis.

Description

Technical field [0001] The present invention relates to the field of occupational health auxiliary diagnosis, and is specifically a method of discriminating an abnormality of an EN binding MPA-SVM based on correlation analysis. Background technique [0002] Underground mining is a very important coal mining method, subject to environmental and equipment under the coal mine, and the physical health of the miner underground cannot be ignored. Dust, chemical poisons in coal mine work environments, harmful physiological factors will affect the health of miners. Between various signs of the human body is interdependent, when the basic body of the human body occurs abnormal, the body of the human body has changed, and these abnormal signs will be a precursor to suffering from the patient. Only a variety of signs parameters can be accurately judged to the health of the human body. [0003] As the efficiency and accuracy requirements of diagnosis continue to improve, the artificial intel...

AI Technical Summary

Problems solved by technology

However, most conventional artificial intelligence algorithms are based on full-attribute data sets, which often contain redundant and useless information, which will affect the final discrimination results, resulting in low diagnostic efficiency and low diagnostic accuracy.

Images