Type-2 Diabetes Dataset Bangladesh

Out of all participants, 840 were living with diabetes, while 225 did not have the condition, giving the dataset a natural imbalance that reflects real-world clinical trends. For each patient, a mix of demographic, lifestyle, and clinical measurements was recorded.

These include the number of pregnancies, age, body mass index (BMI), blood pressure (systolic and diastolic), diabetes pedigree function (a measure of family history), insulin levels, skin thickness, and glucose levels. The dataset also clearly labels whether the patient was diabetic or non-diabetic, making it suitable for supervised machine learning tasks. Because it combines medical history, physical measurements, and biochemical tests, this dataset can be used in many ways: Building and testing machine learning models for diabetes prediction Exploring risk factors and feature importance Developing early detection tools Supporting epidemiological studies on diabetes in diverse populations Highlights: Patients: 1,065 Diabetic: 840 | Non-diabetic: 225 Age range: 21–86 years Variables: demographic, clinical, and biochemical features Target: Type-2 Diabetic (Yes/No)