Abstract :

Hospital Readmission, Defined As The Re-hospitalization Of A Patient Within 30 Days Of Discharge, Remains A Major Challenge In Modern Healthcare Systems. It Imposes A Significant Financial Burden On Healthcare Providers And Patients While Also Indicating Potential Gaps In The Quality Of Care And Patient Recovery. Studies Show That Nearly One In Five Patients Is Readmitted Within 30 Days, Leading To Billions Of Dollars In Avoidable Healthcare Costs Annually. This Project Aims To Develop A Machine Learning-based Predictive System To Identify Patients At High Risk Of Hospital Readmission. The Model Is Built Using The Hospital Readmissions Dataset From Kaggle, Which Includes Over 25,000 Patient Records With Features Such As Age, Time Spent In Hospital, Number Of Lab Procedures, Medications, Diagnoses, And Diabetes Status. Various Supervised Machine Learning Algorithms, Including Logistic Regression, Random Forest, K-Nearest Neighbors (KNN), Support Vector Classifier (SVC), Naive Bayes, Neural Networks, And LightGBM, Are Implemented And Evaluated. The Models Are Assessed Using Performance Metrics Such As Accuracy, Precision, Recall, And AUC-ROC Score. Experimental Results Indicate That The Neural Network Model Achieves The Highest Accuracy Of Approximately 87.66%, Followed By LightGBM (87.28%) And Random Forest (86.54%), Demonstrating Strong Predictive Capability. Keywords: Hospital Readmission, Machine Learning, Predictive Analytics, Healthcare, Neural Networks, Random Forest, Classification, EHR, Patient Outcomes

Published:

06-4-2026

Issue:

Vol. 26 No. 4 (2026)

Page Nos:

1241-1248

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite