Diabetes is ranked among the most severe diseases globally. It results from a several factors such as being overweight, high blood sugar levels and additional reasons. This happens by changing the insulin hormone leading to an abnormal metabolism in the body and increasing the levels of sugar in the blood. Uncontrolled diabetes can raise the risk of other health problems such as cancer, kidney failure and vision loss. This research proposes carrying out a comparative study to tackle the mentioned issues depending on categorization and forecasting methods. Machine and deep learning methodologies are currently under investigation to create smart and effective systems for detecting diabetes. In this study categorization methods are applied and the data is collected from the Pima Indian Diabetic Database located at the University of California, Irvine's Machine Learning Laboratory. The dataset includes information on younger female patients with the age between 21 and 35 years. In this study, a comparative analysis was carried out to forecast diabetes in women through the application of machine learning classification techniques on a dataset that encompasses Decision Tree T (DT), Support Vector Machine (SVM), and Random Forest (RF). Upon reviewing the statistical data, it has been noted that the RF excels, achieving a peak accuracy rate of 80%.