|
Towards Machine Learning Algorithm for Screening Prediction of COVID-19 Patients |
|---|---|
| รหัสดีโอไอ | |
| Creator | Siwakorn Banluesapy |
| Title | Towards Machine Learning Algorithm for Screening Prediction of COVID-19 Patients |
| Contributor | Waraporn Jirapanthong |
| Publisher | The Association of Council of IT Deans (CITT) |
| Publication Year | 2565 |
| Journal Title | Journal of Information Science and Technology |
| Journal Vol. | 12 |
| Journal No. | 1 |
| Page no. | 47-60 |
| Keyword | Randomforest, Neural network, Naive Bayes, Machine Learning, Covid-19 |
| URL Website | https://tci-thaijo.org/index.php/JIST |
| Website title | Journal of Information Science and Technology |
| ISSN | 2651-1053 |
| Abstract | This research aims to develop aweb-based applicationfor screening COVID-19 patientsby applying withmachine learning techniques. We have applied the techniques to classify the data and then compare thealgorithm'sefficiency in prediction of COVID-19 patient screening. The data on 1,608,923 patients from the Department of Disease Controlare accumulated.In particular, the data fromJanuary 1, 2020 to October 1, 2021 are learnt bythree classificationalgorithms: Random Forest, Neural network, and Naive Bayes.The performance of predictive techniques between featuresare compared. The models learnt and generated based on different three algorithms are tested by the Split Test method.The data arerandomly divided into two parts:Training Data (80%) and Test Data (20%) byusing Orange Canvas.The models are then experimented to determine the performance test results withthe highest accuracy. As the results, it is found that the accuracy rates of Random Forest, Neural Network, and Na๏ve Bayes are 93.33%, 92.7%, and92.1%, respectively. The model based on Random Forest technique with the highest value of accuracy is thenapplied in the application for screening COVID-19 patients. |
