Challenge 6 - Heart Failure Prediction

Challenge 6: Heart Failure Prediction

Level: Medium

Description: You are a medical researcher working with a hospital to uncover key risk factors behind heart failure. Using an unbalanced dataset of patient records, your task is to build a predictive model to identify potential heart disease cases. But accuracy alone isn’t enough—clinicians want to understand why the model makes the predictions it does.Train your model, then apply explainable AI techniques to reveal the top three features influencing its decisions. Can your insights help doctors detect heart failure earlier and more effectively?

Beginner-friendly objective(s): 1. Load and preprocess the heart disease dataset, ensuring that the data is clean and ready for analysis. 2. Perform a train-test split on the dataset, maintaining the class distribution for accurate model evaluation.

Intermediate-friendly objective(s): 1. Implement a parameter optimization loop to fine-tune the model's hyperparameters for improved performance. 2. Within the Parameter Optimization Loop, conduct cross-validation to assess the model's robustness and generalization (default: Naïve Bayes, but feel free to experiment with other models).

Advanced objective(s): 1. Integrate multiple data science techniques, including one-hot encoding and normalization, to enhance the model's predictive power. 2. Evaluate the model's performance using advanced metrics and visualization techniques to gain insights into its accuracy and reliability. 3. Use the Surrogate Random Forest model from Global Feature Importance to determine the top 3 most important features driving predictions.

What are the top 3 features responsible for the model's predictions?

Dataset

The top 3 features responsible for the model's predictions are:

1. ST_Slope

2. ExerciseAngina

3. ChestPainType

These features play the most significant role in determining the likelihood of heart failure based on the trained model.