Predicting Covid-19 based on symptoms using machine learning techniques implemented in Python

The sudden spread of the Covid-19 virus all over the globe and eruption of the infection requires an epidemiological examination of the malady in a short period and boost awareness of worthwhile interventions. In this paper, three machine-learning classification models that predict positive SARS-CoV-2 contamination based on symptoms like fever, cold, tiredness, etc are demonstrated. Covid dataset from Kaggle has been taken for the study. The comparison of different models based on the accuracy of the results given by various models is done. These models are SVM, KNN, and Decision Tree. The dataset has been randomly partitioned in which 80% data is reserved for training and 20% data is used for testing. For the validation purpose in KNN 10-fold cross-validation method is used. All three classifiers give efficient results. The reason for this much accuracy is that first of all the testing is done using supervised learning and secondly, only those attributes of the dataset are taken for model construction which has a high correlation with the target field. The symptoms such as nasal congestion, difficulty in breathing, tiredness are highly correlated with the target field. In this research work, Python programming language which is a very powerful language for machine learning is used.