"Diabetes Disease Classification Dataset"

"This dataset consists of clinical records of patients who were assessed for risk of diabetes. We compiled 2000 plus patient records, recorded at least 19 clinical and demographic features, such as age, sex, fasting and postprandial blood glucose, glycated hemoglobin (HbA1c), body mass index (BMI), random or standard blood pressure, lipid profile, renal function, thyroid\u2011stimulating hormone. Here, the target class is Diabetes_Status, and encoded target class is Diabetes_Target, which is classified into three\u2011classes, where namely as Non\u2011Diabetic, Pre\u2011Diabetic, and Diabetic. The raw data contains a lot of duplicate, NaN, and outliers in physiological ranges, which means that careful preprocessing steps need to be in place. The dataset covers a broad range of metabolic and cardiovascular type complications, thus being appropriate to conduct exploratory analyses and predictive modelling of the progression from diabetes to its associated disease outcomes once appropriate quality control has been conducted."