IDENTIFY CHOLESTEROL DISEASE RISK LEVELS USING MULTIPLE LINEAR REGRESSION ALGORITHMS

Cholesterol is one of the fat compounds found in bloodstream that are necessary for formation several hormones and new cell walls liver. Normal cholesterol levels human body range < 200 mg / dL. If blood abnormal or excessive, it can result dangerous diseases such as heart disease stroke. In this study, prediction will be carried out using models formed from linear regression methods, so results study used a reference early prevention become means decision making. Linear methods data mining to find how dependent variables/criteria predicted through independent variables predictor individually. by utilizing some patients with has been stored database attributes, namely age, BMI, glucose, cholesterol. So applying algorithm done identification based on functional relationships attributes data. The showed an RMSE value 0.347 standard deviation /- 0.000. This shows model resulting algorithms above cases quite accurate.