A Comparative Analysis and Prediction Model for Diagnosing Covid19 Surge on People Having Anxiety by Using Machine Learning Techniques

Due to COVID-19 and pandemic emergency might affect the safety, health and well-being both as individuals (causes insecurity, stigma, emotional isolation and confusion). All communities (results deficient distribution of necessities, economic loss of jobs and school closures, insufficient resources for medical responses). These will effect to translate into a scale of emotional reaction (distress or psychiatric scenarios), lack of health symptoms (extra substance in use), and with noncompliance with public health regulations (such as home confidence and vaccination) in population who influences with stress, depression and anxiety. Addition to this autism ASD(Autism Spectrum Disorder) affected personalities might undergoes with more panic with these unexpected pandemic situations especially with children. Environmental situations that will affects communications and reciprocal social interactions, repetitive behaviors. To classify and predict these types of effected people. This research initially classifies people based on dictionaries (+ve and ‒ve) as keywords with high variance by bagged trees for training observations. The next level is to filter the people based on the dictionaries and their commonalities with respect to distance vector mechanism with mixed attributed data. Dramatically altered the predictions of people by using these classified results in the prediction of learned tree (bagged tree) from support vector machine (SVM) and Random Forest (RF) approaches. With SVM and RF high regression was found in perspective of predicting the people who are suffering from anxiety, stress and depression during this pandemic time by considering with various other parameters.