Comparison of linear discriminant analysis and logistic regression for classification diabetes data in Kirkuk Governate

Diabetes is a chronic metabolic disorder that has yet to be addressed. It's a collection of metabolic disorders that impair protein, fat, and carbohydrate utilisation. These disorders result from a lack of insulin synthesis or a poor tissue response. The number of these illnesses is growing internationally as the prevalence of obesity and the quantity of physical activity decreases, resulting in the development of diabetes. As a consequence, the World Health Organisation has identified obesity as a major concern. In this discipline, categorisation is the key focus of study, and an increasing number of scientists use techniques based on various approaches. This research uses logistic regression (LR), a statistical approach for interpreting complicated data combinations. Discriminant analysis is a technique of observation that categorises objects based on the degree of their dependent variable and then generates a formula that may be used to add additional information. The findings demonstrate that logistic regression of medication, blood pressure, age, gender, waist circumference, and LDL levels was used in the discriminant analysis for diabetes categorisation using a simple logistic regression model. The ROC curve and AUC value of 0.61 show that the logistic regression model can identify between diabetic and non-diabetic individuals.