A Data Science Solution for Analyzing Long COVID Cases

Abstract

Many people around the world have witnessed various repercussions caused by the COVID-19 pandemic, such as a decline in industrial activities and business closures. A notable negative consequence of this situation is the potential impact of long COVID on workers across multiple industries, particularly in the industrial sector. As significant volumes of data have been collected during both the COVID-19 period and the subsequent post-COVID-19 period, researchers have initiated investigations into the condition commonly known as long COVID. In this paper, we present a data science solution that integrates data from diverse and comprehensive sources to uncover meaningful associations within demographic data related to long COVID. Leveraging this integrated information, our solution identifies features leading to long COVID in patients. Evaluation results on real-life datasets demonstrate practicality of our solution in identifying individuals who may be prone to long COVID, while also highlighting demographic factors that may indicate an elevated risk. Through evaluation, we show the practicality of our solution in analyzing and predicting long COVID cases.