Type 2 diabetes occurs more commonly in middle-aged and elderly people and if left uncontrolled it can cause all sorts of serious health issues like infections, damaged kidneys, vision loss and blindness, amputations and many more. So, there is no question that type 2 diabetes needs to be taken seriously and treated.
Type 2 diabetes is usually diagnosed using the glycated hemoglobin (A1C) test. In this project we will try to predict A1C levels: no-diabetes (below 5.7), pre-diabetes (5.7 to 6.4), diabetes (6.5 or higher). We will transform the dataset from a regression task (A1C) into a multi-class classification task (3 A1C levels).
The notebook contents all coding, visualization and detailed description of all steps.
- Tools/techniques used: Python, Jupyter Notebook, Seaborn, Scikit-learn, SMOTE, hyperparameter tuning
- Algorithms used: Logistic Regression, SVM, Random Forest, Gradient Boosting, AdaBoost
GitHub Repository for this project.