Machine-Learning-Diabetes

On the basis of given medical parameters such as blood pressure,glucose and BMI in the dataset, we use machine learning to train and test which patient has diabetes. Three algorithms were employed for this purpose - Random forest, logistic regression and decision tree. We found that logistic regression was the most effective in testing which patients had diabetes with an accuracy of about 79.87%. Also heatmaps and pair plots between variables have been plotted.