Event Details
COVID-19 Prediction Using Supervised Machine Learning
Presenter: Irfan Ali
Supervisor:
Date: Thu, December 22, 2022
Time: 08:00:00 - 00:00:00
Place: ZOOM - Please see below.
ABSTRACT
Zoom meeting link: https://uvic.zoom.us/j/88341558822?pwd=Q0I0d3BJK3hKR1JiQmxZbTJVZUJQdz09
Meeting ID: 883 4155 8822
Password: 986809
One tap mobile
+16475580588,,88341558822# Canada
+17789072071,,88341558822# Canada
Dial by your location
+1 647 558 0588 Canada
+1 778 907 2071 Canada
Meeting ID: 883 4155 8822
Find your local number: https://uvic.zoom.us/u/kcn2r3nqKL
Note: Please log in to Zoom via SSO and your UVic Netlink ID
ABSTRACT:
Early diagnosis is important to stop the spread of illnesses that endanger human life. COVID- 19 is a contagious disease that has mutated into multiple variants and created a global epidemic that requires immediate diagnosis. With the increase in COVID-19 cases, the amount of associated data grows every day, and data mining can be used to extract information from this data. In this project, a COVID-19 symptoms and presence dataset is used with several supervised machine-learning algorithms to predict COVID-19 in the human body by examining the symptoms. The Bayes Net, Simple Logistic, Bagging, Support Vector Machine (SVM), and AdaBoost M1 classifiers are considered using the open-source Waikato Environment for Knowledge Analysis (WEKA) Machine Learning (ML) tool. Principal Component Analysis (PCA) is used to reduce the number of features in the dataset based on eigenvalues. Then the model is trained and tested using 5-fold cross-validation, 10-fold cross-validation, and 66/34 and 34/66 split. The performance of each model is evaluated based on accuracy, precision, recall, F-measure, and execution time. The results obtained show that Bagging outperforms the other classifiers with an accuracy of 99.3% and an execution time of 0.10 s for a 66/34 split using 10 features.