Framework for detecting the patients affected by COVID-19 at early stages using Internet of Things along with Machine Learning approaches with improved Accuracy
T Devi., J. Sathya Priya, N. Deepa
Abstract
COVID-19 has been affecting the entire world from the year 2019 and in order to handle this pandemic situation, it is necessary to follow the measures till valid medicine has been found. The proposed approach helps in detecting as well as monitoring the COVID-19 on real time basis. Data is collected with the help of IoT devices for detecting this disease at the early stage. The components of the system are, (i) Collection of symptom data, (ii) Center of Isolation, (iii) Machine Learning approaches for analysis, (iv) Healthcare analysts and (v) Cloud. The three algorithms in machine learning used for detection of the virus are Decision tree, Support vector machine and Neural Network. These three algorithms are tested with the real time dataset and it is observed that all these algorithms have accuracy greater than 91%. Identifying the disease accurately with the three machine learning algorithms, effective results are produced. The response of treatment for every person who gets affected by the virus is then documented.