Evaluation of Public Hospitals' Performance with Decision Tree Algorithms

Purpose: The study aims to evaluate a range of financial performance indicators calculated through structural, operational, and HVI measures for public hospitals in the Turkish healthcare sector using various decision tree algorithms. Methodology: The study comprises threa phases. In the first phase...

Full description

Saved in:
Bibliographic Details
Main Author: Keziban Avcı
Format: Article
Language:English
Published: Sanayi ve Teknoloji Bakanlığı 2025-01-01
Series:Verimlilik Dergisi
Subjects:
Online Access:https://dergipark.org.tr/tr/download/article-file/3974713
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Purpose: The study aims to evaluate a range of financial performance indicators calculated through structural, operational, and HVI measures for public hospitals in the Turkish healthcare sector using various decision tree algorithms. Methodology: The study comprises threa phases. In the first phase, financial ratios were calculated from the hospitals' financial statements using the ratio analysis method. In the second phase, these ratios were used to calculate the HVI. In the third phase, the selected operational and financial indicators were analyzed with decision tree algorithms. The ID3, C4.5 and CART decision tree algorithms and AUC were used for predicting operational and financial indicators and performance assessment of decision trees.Findings: It has been observed that decision trees created using the ID3 algorithm exhibit higher performance compared to other algorithms (AUC = 0.93). According to the results of the study, the number of beds significantly predicts the operational and financial performance of public hospitals and can be explained by the hospital scale. In addition, a strong relationship was found between operational and financial performance indicators with training status.Originality: The study is original in demonstrating the effectiveness of the ID3 decision tree algorithm in predicting the performance of public hospitals.
ISSN:1013-1388