Prediction of Impairment Biomarkers in Smoker Based on Machine Learning

Smoking is one of the major health issues globally, causing up to 7 million deaths each year. This paper aimed to predict impairment biomarkers in smoker based on machine learning. Firstly, we selected data set GSE37768 from the GEO database. And then, we standardized and removed batch effects from the data set. Thirdly, we select some genes through LogFC and significance tests. Lastly, the combination of 5 machine learning methods(support vector machine, random forest, decision tree, logistic regression, naive Bayes) and recursive feature elimination-cross validation was used to prediction impairment biomarkers. We found that identified biomarkers were PLSCR4, CDH3, RUNX1T1, and GUCY1A2.