Scholarlink

Yellow fever, a febrile disease, remains a significant public health concern, especially in low- and middle-income countries (LMICs) in Africa, where its prevalence is driven by a combination of socioeconomic and environ-mental factors. This study explores the application of machine learning (ML) techniques to enhance the diagnosis of yellow fever. The dataset comprised over 4,870 patient records obtained from secondary and tertiary healthcare facilities in the Niger Delta region of Nigeria, with contributions from 62 experienced physicians specializing in febrile illnesses. Four classifiers, Naïve Bayes, K-Nearest Neighbor (KNN), RIPPER, and PART, were employed to assess their effectiveness in predicting yellow fever cases using causative data as opposed to clinical symptomatic data. The four models demonstrated similar performance in accuracy, sensitivity and precision indicating their strength in accurately identifying true yellow fever cases, which is critical for timely intervention and treatment. This study will benefit the different public stakeholders as it underscores the potential of ML models to improve the accuracy of yellow fever diagnoses in LMICs.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Comparative Analysis of Machine Learning Classifiers for Yellow Fever Diagnosis Using Causative Data: Evaluating Naïve Bayes, KNN, RIPPER, and PART

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Said Baadel,
  • Kingsley Attai,
  • Brain Bassey,
  • Ekerette Attai,
  • Asim Majeed,
  • Faith-Michael Uzoka

摘要

Yellow fever, a febrile disease, remains a significant public health concern, especially in low- and middle-income countries (LMICs) in Africa, where its prevalence is driven by a combination of socioeconomic and environ-mental factors. This study explores the application of machine learning (ML) techniques to enhance the diagnosis of yellow fever. The dataset comprised over 4,870 patient records obtained from secondary and tertiary healthcare facilities in the Niger Delta region of Nigeria, with contributions from 62 experienced physicians specializing in febrile illnesses. Four classifiers, Naïve Bayes, K-Nearest Neighbor (KNN), RIPPER, and PART, were employed to assess their effectiveness in predicting yellow fever cases using causative data as opposed to clinical symptomatic data. The four models demonstrated similar performance in accuracy, sensitivity and precision indicating their strength in accurately identifying true yellow fever cases, which is critical for timely intervention and treatment. This study will benefit the different public stakeholders as it underscores the potential of ML models to improve the accuracy of yellow fever diagnoses in LMICs.