Rough Sets in COVID-19 to Predict Symptomatic Cases

Abstract

Rough set theory is a new mathematical or set-theoretical practice to study inadequate knowledge. There are many use cases in the real world where there is a lack of crisp knowledge. In view of this, many Scientists have been attempted to address anomalies associated with imperfect knowledge for a long time. In recent times, computer and mathematics researchers have been trying to resolve this decisive issue, mainly in artificial intelligence province. The COVID-19 pandemic encroaches the harmony of the whole world. Many patients of COVID-19 have different symptoms, so it is very difficult to carry out the symptoms-based prediction COVID-19. However, the rough set theory approach help to minimize the number of attributes from the underlined decision table. This work defines the decision table having patients and symptoms of the COVID-19 in the rows and columns respectively. By studying data indiscernibility, elementary sets are specified for each attribute. Moreover, lower approximation, upper approximation, class of rough sets and accuracy of approximation are defined for different individual or group symptoms. This proposed work investigates whether particular symptoms belong to the decision set or not and also the accuracy of observations is calculated and analyzed. The probability of having COVID-19 is defined by considering the different sets of attributes. The main objective of this work is to minimize the number of symptoms of COVID-19 by rough set theory approach for better decision making. This symptoms-based prediction could help us while checking patients and decision-makers could be benefited while making policies and guidelines.