Early Diagnosis Prediction from COVID-19 Symptoms Using Machine Learning Methods
Knowledge Commons (Lakehead University) · 01.01.2024
DOI: 10.17613/4g2cr-kav81 ↗Kurz gefasst
Fachartikel in der Zeitschrift Knowledge Commons (Lakehead University), erschienen am 01.01.2024. Verfasst von Charlyn V. Rosales.
Abstract (Originalsprache)
This study proposes a new method for detecting COVID-19 using Artificial Neural Networks by analyzing a person's current symptoms without requiring laboratory tests. The methodology utilized in this study includes the data collection of the dataset from Kaggle, then the Exploratory Data Analysis was performed for data pre-processing to attain a clean and comprehensive dataset to be used to train the prediction model. To determine the highest possible performance of the algorithm, GridSearchCV was used for hyperparameter tuning and 10-fold cross validation to optimize Artificial Neural Networks performance then a prediction model was developed using the optimal configuration. The results suggest that hidden layer sizes of (100,), (50, 100, 50), and (50, 50, 50), relu and tanh activation functions, adam solver, 0.05 and 0.0001 alpha values, and adaptive and constant learning rates were the values that achieved the best algorithm performance. Experimental results show that the developed prediction model attained 98.86% accuracy, 98.79% specificity, and 100% sensitivity. This prediction model can be utilized in applications that integrate prediction models to determine the presence of the COVID-19.
Schlagwörter
Zitieren
APA
Rosales, C. V. (2024). Early Diagnosis Prediction from COVID-19 Symptoms Using Machine Learning Methods. Knowledge Commons (Lakehead University). https://doi.org/10.17613/4g2cr-kav81
Vancouver
Rosales CV. Early Diagnosis Prediction from COVID-19 Symptoms Using Machine Learning Methods. Knowledge Commons (Lakehead University). 2024. doi:10.17613/4g2cr-kav81
BibTeX
@article{Rosales2024,
author = {Rosales, Charlyn V.},
title = {Early Diagnosis Prediction from COVID-19 Symptoms Using Machine Learning Methods},
journal = {Knowledge Commons (Lakehead University)},
year = {2024},
doi = {10.17613/4g2cr-kav81}
}
EUVIMED