ABSTRACT
The main focus of this Research is to see how one can easily predict if he/she has been infected by COVID-19. Another aspect is deciding which COVID-19 symptom is more likely to show positive result of virus contamination. The virus has been declared as a pandemic and has affected more than 66,729,375 people across 220 countries and has also cost the lives of 1,535,982 people [source : who.int] as of the time this paper is being written. Research still predicts that another second wave is to hit soon. The required objective is obtained using complex machine learning algorithms that are able to predict, up to an extent the probability that the person has covid-19 and also if the related covid-19 symptoms provided are relevant to the condition or not. The algorithm used is a LOGISTIC REGRESSION algorithm that is used as a classification tool to separate the data into binary results, which in our case is if the person has covid-19 or not (YES OR NO). The dataset used to train this machine learning algorithm is obtained from online resources and a public survey.The machine learning model as of now has been able to predict the probability of virus contamination by 66.89% accuracy, further the model is able to relate if a given symptom is valid or not. With this we are able to conclude that the model is working fine and only fine tuning of the model is required in order to improve and enhance the accuracy and probability of exact results. The result were then improved using ANN( Artificial Neural Networks) but as it is computational expensive and due to lack of resources the expected performance cannot be met. The maximum accuracy achieved with ANNs was about 71%. © 2021 IEEE.