| Abstrakt: |
The adaptation of Artificial Intelligence can radically reshape the entire healthcare industry. This paper proposes a comparative analysis of four Machine Learning algorithms namely, k-Nearest Neighbour, Naive Bayes, Decision Tree, and Random Forest. These supervised classifiers are used to predict widely identified diseases based on one’s conspicuous symptoms among a given data-set of common symptoms. In our experiment comparing the different Machine Learning models, Random Forest had obtained the highest accuracy of 99.5%, followed by Decision Tress at 95.8%, K-Nearest Neighbor at 93.4%, and Naive Bayes at 87.7%. This paper represents a state-of-the art comparative analysis which has achieved the higher accuracy than the existing analytical results carried out in earlier research. Finally, a web-based application has also been developed to visualize the predictions in a better way. |
Full text from SpringerLink