Selecting the Best Health Care Systems: An Approach Based on Opinion Mining and Simplified Neutrosophic Sets

Abstract

Measuring what hospital offers the best services is very difficult, for that reason, the opinions from previous patients have become an essential tool for the new possible clients to decide which services they must select. Many online platforms deal with opinions to analyze their services/products, primarily, by means of aspect-based sentiment analysis techniques. These techniques are mainly based on the detection of features from services/products to model the feelings toward them. Most models primarily cope with positiveness, negativeness and neutrality; nevertheless, these do not reflect other situations in which there are positive and negative aspects, but the overall sentiment is not neutral, but indeterminate. To face this issue, simplified neutrosophic sets can be a useful tool. Therefore, this study presents a novel application of the simplified neutrosophic sets to hospital ranking. The application, first, detects the hospital features, second, models the feelings toward them using simplified neutrosophic sets, and finally, ranks the hospitals according to the patients’ preferences. It has been tested using opinions from a real website and compared against other fuzzy logic-based approaches. The achieved results outperform the ones obtained by other proposals, revealing that simplified neutrosophic sets can be an interesting solution to model feelings.