This project aims to develop a machine learning model for predicting chronic kidney disease (CKD) using the Fuzzy K-Nearest Neighbors (Fuzzy KNN) algorithm. Chronic kidney disease is a significant health concern globally, and early detection plays a crucial role in effective management and treatment. Machine learning techniques, particularly Fuzzy KNN, offer a promising approach for accurate prediction.
- numpy
- pandas
- matplotlib
- seaborn
- scikit-learn
- scipy
Dataset Source : Kaggle
| Attribute Name | Actual Name | Attribute Name | Actual Name |
|---|---|---|---|
| age | Age | sod | sodium |
| bp | blood pressure | pot | potassium |
| sg | specific gravity | hemo | hemoglobin |
| al | albumin | pcv | packed cell volume |
| su | sugar | wc | white blood cell count |
| rbc | red blood cells | rc | red blood cell count |
| pc | pus cell | htn | hypertension |
| pcc | pus cell clumps | dm | diabetes mellitus |
| ba | bacteria | cad | coronary artery disease |
| bgr | blood glucose random | appet | appetite |
| bu | blood urea | pe | pedal edema |
| sc | serum creatinine | ane | anemia |
| pcc | pus cell clumps | class | class |
- Data Preprocessing: Cleaning the dataset, Changing Attribute Name, handling missing values, and encoding categorical & nominal.
- Feature Selection: Identifying relevant features that contribute significantly to the prediction of CKD.
- Model Training: Implementing the Fuzzy k-NN algorithm.
- Model Evaluation: Assessing the performance of the model using appropriate metrics such as accuracy, precision, recall, and F1-score.
- Clone the repository :
git clone https://github.com/DikkiKartajaya/ChronicKidneyDisease-FuzzyKNN.git
- Install the required dependencies :
pip install -r requirement.txt
- Run the Jupyter notebook Kidneydisease_FuzzyKNN.ipynb to train and evaluate the Fuzzy KNN model.
Contributions to the project are welcome! If you have any suggestions for improvement, feature requests, or bug reports, please feel free to open an issue or submit a pull request.