Hospital readmission is a critical issue in the management of diabetic patients, resulting in significant financial burdens on healthcare systems and highlighting potential lapses in patient care. This project aims to develop a predictive model to assess the risk of hospital readmission for diabetic patients, enabling healthcare providers to take timely interventions and reduce avoidable readmissions.
Traditional approaches to feature selection and classification, such as Chi-square analysis, may overlook complex relationships between features and readmission risk. In this project, we introduce novel enhancements to improve the accuracy of readmission risk prediction models, ultimately benefiting both patient outcomes and healthcare resource management.
We tested and optimized various machine learning classifiers using the techniques mentioned:
- Decision Trees
- Random Forests
- KNN
- Logistic Regression
- XGBoost
- Dataset Link: https://archive.ics.uci.edu/dataset/296/diabetes+130-us+hospitals+for+years+1999-2008
- Refer to Report for Methodology.
Contributions are welcome! Please fork this repository and submit a pull request for any enhancements or bug fixes.
This project is licensed under the MIT License.