Machine Learning Algorithms for IoT based Covid-19 Patients Management System
| dc.contributor.author | Muhammad Umar Ehsan | |
| dc.date.accessioned | 2025-09-24T15:43:21Z | |
| dc.date.available | 2025-09-24T15:43:21Z | |
| dc.date.issued | 2023 | |
| dc.description.abstract | Coronavirus disease (COVID-19) is an infectious disease caused by the SARS-CoV-2 virus. More than 548 million positive cases and 6.34 million deaths have been reported worldwide. During the time of pandemic hospitals were flooded with patients and paramedical staff wasn’t enough for patients. In this study we have proposed a machine learning approach to build a smart healthcare system for detecting critical covid-19 patients so they could be moved into intense care units (ICUs) automatically. For the experiment internet of things (IoT) based sensors data is used to build a machine learning model, first 40 machine learning algorithms were tested on the dataset, after that top 3 classifiers were selected. To check the model’s robustness K-fold cross validation testing is performed on the dataset for the top 3 classifiers in which every part of the dataset is used to train and test the algorithm, after that mean is calculated for the accuracies. In this experiment Extra Trees classifier has achieved 92% accuracy after K-fold cross validation testing. | |
| dc.identifier.uri | https://escholar.umt.edu.pk/handle/123456789/6871 | |
| dc.language.iso | en | |
| dc.publisher | UMT, Lahore | |
| dc.title | Machine Learning Algorithms for IoT based Covid-19 Patients Management System | |
| dc.type | Thesis |
Files
Original bundle
1 - 3 of 3
No Thumbnail Available
- Name:
- CamScanner 03-22-2023 00.07.pdf
- Size:
- 142.28 KB
- Format:
- Adobe Portable Document Format
No Thumbnail Available
- Name:
- Machine Learning Algorithms for IoT based Covid-19.pdf
- Size:
- 1.67 MB
- Format:
- Adobe Portable Document Format
License bundle
1 - 1 of 1
No Thumbnail Available
- Name:
- license.txt
- Size:
- 1.71 KB
- Format:
- Item-specific license agreed upon to submission
- Description: