A NOVEL FEATURE SELECTION METHOD FOR PREDICTION OF FACTORS AFFECTING ANXIETY IN HEALTH CARE WORKERS DURING PANDEMIC

Abstract

During the pandemic, hospital workers are more likely to report symptoms of sadness, anxiety, and stress. The objective of this study is to determine the elements that cause anxiety in pandemic among health care workers (HCWs). Survey comprising of latest and relevant articles stating the anxiety factors among HCWs of Saudi Arabia and rest of the remaining articles comprising of other countries stating anxiety factors of HCWs in their respective region were carefully analyzed and recorded. Moreover, a benchmark health care dataset is utilized and evaluated through proposed feature selection technique. Identified factors by proposed technique are than compared with the reported factors causing anxiety in health care workers of Saudi Arabia and other countries are analyzed. Chi-square, F_Chi and mRMR feature selection techniques are utilized to predict the factors affecting the anxiety of health care workers. The results enlisted the number of factors that contributed in accelerating anxiety among HCWs. Those factors were further applied for Classifiers like NC, KNN, AB and GB has low accuracy, precision and recall scores. However, classifiers namely MLP, NB, SGD and SVM has better overall accuracy scores. Among all, SVM Classifier stands out the most with Accuracy score of 91%, moreover 90% scores in Precision and recall.