Performance Analysis of Machine Learning Algorithms for Disease Prediction

Priya, B. H.; Chaitra, C.; Reddy, K. V.

With the recent technological advances in microelectronics, wireless communication, machine learning (ML), and decision-making process, Wireless Body Area Network (WBAN) has become the most promising technology. As we all know that we are in global pandemic due to Covid-19 situation now, hence, there is a demand occurring in health care services and continuous monitoring. Moreover, prediction of abnormalities at an early stage will be crucial for a person in diagnosis. Hence, in this paper we have developed and compared the performance of three machine learning algorithms such as Decision Tree Classifier (DTC), K-Nearest Neighbor (KNN), and Random Forest (RF). Each algorithm is tested with datasets of 100, 200, 500 & 1000 users respectively. Further, threshold values have been identified by consulting with doctors for accurate disease prediction based on the vital signals collected by various sensors. The three algorithms used are based on supervised learning, where the output is predicted based on the training of the developed classifier. From the results, it is observed that the accuracy in disease prediction using RF is 0.99 & outperformed when compared with state of the art for datasets of 1000 users. © 2021 IEEE.