Myopia prediction and early warning model based on naive Bayesian model
Download as PDF
In recent years, the problem of myopia in children and adolescents in China has been significantly intensified and approached to a younger age. This paper analyzes the causes of myopia and the early warning conditions of myopia by establishing a reasonable mathematical model. After collecting data, we obtained age, genetic factors, eye time for over-loading, education level, and sleep time as the main influencing indicators. Different quantitative models were established through a large number of statistical laws, and the eyeball structure was studied. With the formula and the lens, The ciliary body-based regulation mechanism and the characteristics of true and false myopia, we have summarized the evolution mechanism model of human myopia. Through Principal Component Analysis (PCA) and SPSS factor analysis, the weights of each feature on myopia are obtained. The features are brought into the target equation to solve the predicted value Y, and the Y* (standard value) calculated by the survey and brought into the model. For comparison, the suitable solution for the teenager is proposed according to different deviations.
Myopia evolution mechanism, naive Bayes classifier, principal component analysis, statistical regression analysis, Gaussian distribution