Automated System to Diagnose and Detect the Allergy, Common Cold, Flu, and Covid using Machine Learning Approaches

Rapid detection based on clinical symptoms from several illnesses with symptoms similar to COVID-19 became crucial for preventing the spread of the pandemic once COVID-19 emerged as a global health concern. In this study, we have presented an approach to classify and predict the four diseases: allergy, covid, flu, and common cold, which share similar symptoms. In this experiment, we have compared the performance of the decision tree, K-nearest neighbors, random forest, support vector machine, linear regression, and Naïve Bayes algorithms. To get the best performance of these algorithms, hyperparameters are tuned with the help of the grid search algorithm. Experimental results suggest that the performance of all algorithms has been improved after hyperparameter tuning.