Article Preview
TopIntroduction
SARS-CoV-2 (severe acute respiratory syndrome coronavirus-2), also known as COVID-19 was initially discovered in Wuhan, China, in December 2019(Zoabi et al., 2021). Since then, it has spread around the world and has affected more than 200 countries. According to the figures of WHO, a total of 239,437,517 infected cases and 4879235 deaths are recorded as of October 15th, 2021(WHO Coronavirus (COVID-19) Dashboard). On March 11th, 2020, the World Health Organization (WHO) designated it a “global pandemic” (Coronavirus Disease (COVID-19)). The United States of America (US), the United Kingdom (UK), Brazil, Russia, and India are the top five nations with the most infected COVID-19 cases worldwide. The spread of COVID-19 is unstoppable, and many new forms of it have emerged that are known to be more infectious. Scientists designated one of them as “VUI - 202012/01”(Mishra, 2020).
COVID-19 is a highly contagious disease that is spread from person to person through droplets created by infected people when they sneeze or cough. Since it is a novel virus, scientists and epidemiologists still have a lot to learn about it. Even though vaccines are now available across countries and people are getting vaccinated, they are getting infected by COVID-19 even after vaccination. Hence to curb the spread of this virus, it is important that it is detected at an early stage. Mass testing is one of the most efficient methods for early diagnosis of this virus. It is, however, not a viable solution due to the large population and insufficient medical facilities. Therefore, the use of technology in the medical field can be another way to control the COVID-19 spread. The Recommender System (RS) (G. Jain et al., 2013b) is one such technique of data mining that can be a cost-effective alternative in the early detection of disease. Traditionally an RS collects the user's historical behaviour, analyses interests, and suggests items according to their interests. Developing a recommender system aims to provide personalized recommendations to customers while selecting an item among a set of products (movies, books, etc.). It became an integral part of the e-commerce business as it has the capability to suggest relevant items to a user depending upon their preferences. Although the idea emerged with the e-commerce domain, it is used in different domains to meet the needs of the users with time (G. Jain et al., 2013a). The health sector is one such domain where RS can be very useful for the primary diagnosis of a disease or to provide users with helpful recommendations based on the symptoms. The system can be based on the patient's profile or symptoms’ profile (Centers for Disease Control and Prevention). The RS's CF technique (G. Jain et al., 2020) is the most popular and commonly used neighbourhood-based technique that considers only user-item ratings and ignores the other details. In CF, a similarity measure is used to determine the relationship between users/items, so selecting the proper similarity measure is important in this technique.