https://github.com/iotchulindrarai/iotchulindrarai-heart-disease-prediction-using-ml-project
IotchulindraRai/Heart-disease-prediction-using-Ml-project with GUI
https://github.com/iotchulindrarai/iotchulindrarai-heart-disease-prediction-using-ml-project
heart-disease heartdisease-prediction machine-learning prediction randon-forest
Last synced: 3 months ago
JSON representation
IotchulindraRai/Heart-disease-prediction-using-Ml-project with GUI
- Host: GitHub
- URL: https://github.com/iotchulindrarai/iotchulindrarai-heart-disease-prediction-using-ml-project
- Owner: IotchulindraRai
- Created: 2023-02-10T09:25:17.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2025-01-26T06:32:29.000Z (9 months ago)
- Last Synced: 2025-06-18T22:40:44.344Z (4 months ago)
- Topics: heart-disease, heartdisease-prediction, machine-learning, prediction, randon-forest
- Language: Jupyter Notebook
- Homepage:
- Size: 131 KB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# IotchulindraRai-Heart-disease-prediction-using-Ml-project
IotchulindraRai/Heart-disease-prediction-using-Ml-project with GUI
training models and make prediction based on model of dataset it was done for learning
linked in : https://np.linkedin.com/in/chulindra-rai-a51308206
🎯 Building a Machine Learning Model to Predict Heart Disease
✅ Dataset: The heart.csv file contains features like age, cholesterol, resting blood pressure, and more to predict the presence of heart disease (output).
🚀 Steps Taken:
1️⃣ Data Cleaning: Removed duplicates and ensured no missing values.
2️⃣ Feature Engineering:
Separated categorical (e.g., chest pain type) and continuous (e.g., age) variables.
Applied one-hot encoding for categorical variables.
Scaled continuous features using StandardScaler.
3️⃣ Model Training: Trained 6 different algorithms:
Logistic Regression (57% Accuracy)
SVM (80%)
KNN (78%)
Decision Tree (78%)
Random Forest (Best! 88%)
Gradient Boosting (77%)
4️⃣ Model Comparison: Visualized results with a bar plot to identify the top performer: Random Forest.
5️⃣ Deployment:
Saved the model using joblib.
Faced feature mismatch issues during new data predictions (fix needed!).
💡 Key Takeaways:
Preprocessing and feature engineering play a huge role in model performance.
Random Forest is a robust algorithm for this problem.
Consistency in feature names is crucial for smooth deployment.
GUI with final output :
