Heart Disease Prediction

Heart disease prediction using machine learning involves building a predictive model that can accurately identify the likelihood of a person having heart disease or not based on various health metrics and risk factors.

Steps Involved

• Data Collection
• Data Preprocessing
• EDA
• Scaling the Data Using MinMaxScalar
• Model building
• Model Evaluation
• Hyperparater Tuning (got 94.21% accuracy with RandomForestClassifier)
• Saving the model using pickle
• Created Streamlit app
• Uploaded to Github

Installation

Install all project requirements to run this project locally

  pip install -r requirements.txt

Run the command to see the demo

  streamlit run app.py

Authors

• @Debopam-Pritam2014

Lessons Learned

• Data Preprocessing: The importance of data cleaning and handling missing values. How to use MinMaxScaler for normalization to ensure that features contribute equally to the model performance.

• Algorithm Selection: Comparison of various machine learning algorithms to find the most suitable one for your dataset. Understanding the strengths and weaknesses of algorithms like Random Forest, SVM, and Logistic Regression.

• Hyperparameter Tuning: The impact of hyperparameters on the performance of the model.Techniques for hyperparameter tuning, such as GridSearchCV, to optimize model performance.

• Model Evaluation: Different metrics to evaluate the performance of the model, such as accuracy, precision, recall, and F1-score. The importance of cross-validation to ensure the robustness of the model.

• Building Applications: How to create a web application using Streamlit to make your model accessible to users. The basics of user interface design to create an intuitive and user-friendly application.

• Version Control: The importance of using GitHub for version control and collaboration. How to document your project and make it accessible to others.

Challenges I Faced and How I Overcame Them

• Feature Selection: Identifying which features are most important for predicting heart disease. Used feature importance scores from the Random Forest model to select the most impactful features.

• Model Overfitting: The model performed well on training data but poorly on test data. Implemented cross-validation and regularization techniques to prevent overfitting.

• Hyperparameter Tuning: Finding the optimal hyperparameters was time-consuming. Automated the process using GridSearchCV to systematically search for the best parameters.

• Deployment Issues: Integrating the model into a Streamlit application and ensuring it runs smoothly. Iterative testing and debugging to resolve integration issues and ensure the application works seamlessly.

Acknowledgements

• Dataset link kaggle
• EDA

Feedback

If you have any feedback, please reach out to me at [email protected]