Classification of Covid-19 Survillance Datasets using the Decision Tree Algorithm

Abstract

Covid-19 is a new type of mutated virus that has been discovered and studied throughout the world. For the time being, no effective drug has been found to treat or prevent this disease. One way that governments around the world are doing is limiting physical contact with people with COVID-19. Data mining is a computer science to study data and perform extraction to get new knowledge. One technique in data mining is classification. C45 is one of the best classification algorithms. The result of the c45 algorithm can be a decision tree. Decision trees are used because the results can be well represented. and can be easily understood in human language. This study classified the Covid-19 surveillance dataset using the Decission tree. The Covid-19 surveillance dataset was obtained from a public data portal, namely the UCI machine learning repository. This study resulted in better accuracy than previous studies using the same dataset. The level of accuracy obtained by using the decision tree algorithm is 65%. Although in this study the accuracy value has increased by 10%, the level of accuracy is still relatively low. The low level of accuracy is due to the dataset used only has 7 attributes and 14 records