Here, some of the classification ML algorithms included in the scikit-learn library are tested.
These ML algorithms are tested in a binary classification problem to predict whether a patient has diabetes or not.
The dataset used in the binary classification problem is the Pima Indians Diabetes Database from the UCI Machine Learning Repository (accessed via Kaggle: https://www.kaggle.com/datasets/uciml/pima-indians-diabetes-database).
The hyperparameters of the classification algorithms are fine-tuned. Additionally, the classification algorithms are used in ensemble models (Voting classifier and Stacking classifier) to evaluate whether an improvement in performance can be achieved.
Finally, a SHAP analysis is performed to get insights into how the model interprets the data and explains the output.
The notebook should be rather self-explanatory and my hope is that it can be helpful for people (like me) who are curious about the field of ML.
I hope you enjoy and get some inspiration.