A Hybrid Approach: Rough Set, Artificial Neural Networking & Data Envelopment Analysis on patient's satisfaction in healthcare organization

Abstract

Data mining plays a significant role in information acquisition and effective information utilisation from big data. Many techniques are available for data mining, In many disciplines, including health care services, data reduction using rough set theory is a commonly utilized mathematical technique. In this paper, Rough Set Theory (RST) is applied to a patient's satisfaction survey to identify sets of critical attributes which are responsible for the patient's satisfaction. RST is a useful tool for data mining, however the data it extracts lacks efficiency and precision. This study proposes a novel approach to tackle this problem, combining Rough Set Theory, Data Envelopment Analysis (DEA), and Artificial Neural Networks (ANN). The Data Envelopment Analysis (DEA) is the leading approach used to find the dependency of output variables on input variables. These powerful instruments Through the use of Artificial Neural Network (ANN) results, RST and DEA are utilised to determine the effectiveness of reductions. Based on cross-validation of ANN accuracy of forecasting is determined. Maximizing patient satisfaction is an important goal of Health care organizations. Patients' feedback helps them to identify the ways to improve their working methods which transforms into better care and happier patients. In this paper, Rough Set Theory is applied to a patient's satisfaction survey to identify sets of critical attributes which are responsible for the patient's satisfaction. DEA & ANN give the best set of critical attributes based on their efficiency.