Scroll Top

Datasets

  • ZADA Dataset (Diabetes)
    The ZADA Diabetes Dataset is a curated medical dataset consisting of health-related features collected from approximately 7,000 patients. After preprocessing steps including data cleaning, integration, and dimensionality reduction, the most relevant features associated with diabetes were selected. The final dataset contains 909 records and seven features, including a binary target variable “Class” that indicates diabetes status (0 = healthy, 1 = diabetic). Data was collected from Shaker Laboratory in Zakho City, located in the Kurdistan Region of Iraq. Key characteristics of the dataset are summarized in the table below.

    Attribute Name Attribute Description Min Max Mean
    Age Age of patients 20 86 48.01
    Cholesterol Test of Cholesterol 110 340 200.56
    L_HDL High-density Lipoprotein 23 65 42.97
    L_LDL Low-density Lipoprotein 36.8 266.2 124.87
    L_VLDL Very Low-Density Lipoprotein 8.6 80 32.73
    Uric Acid Test of Uric Acid 2.22 10.2 5.72
    Class 1= Positive and 0= Negative