A method for diagnosing the disease of Covid-19 based on the Trees Social Relations Optimization Algorithm and Naive Bayes classifier

Articles in Press

Document Type : Original Article

Authors

1 Department of computer engineering, Rasht Branch, Islamic Azad University, Rasht, Iran

2 Islamic Azad University-Rasht Branch

Abstract

Covid-19, commonly known as coronavirus, is a viral disease caused by the SARS-CoV-2 virus. Common symptoms of this disease include fever, cough, feeling tired and loss of sense of smell. The standard method for accurate diagnosis of Covid-19 is the rRT-PCR test, which requires breath sampling, which is time-consuming. Therefore, the development of rapid diagnostic methods for this disease is very important. In this article, a new method for diagnosing Covid-19 using artificial intelligence is introduced. This method, which is fast and non-invasive, is designed based on Tree Social Relationship (TSR) algorithm and Naive Bayes classification. The proposed method includes two main stages of feature selection and disease diagnosis. Selection of features is done using TSR algorithm and disease diagnosis is done by Naive Bayes classifier. The proposed method was practically tested with the COVID-19 Dataset. Practical evaluations have shown that this new method performs better than other existing methods in diagnosing Covid-19 and diagnoses this disease on average with 96% accuracy, 97% recall and 96% F1-score.

Keywords

Main Subjects

Soft Computing Journal

Articles in Press, Accepted Manuscript
Available Online from 10 March 2025

Files

History

  • Receive Date: 23 April 2024
  • Revise Date: 02 March 2025
  • Accept Date: 07 March 2025

Share

How to cite

Statistics