Article: Prediction of dengue fever using supervised machine learning based on hyperparameter tuning and an analysis of the factors influencing dengue spread Journal: International Journal of Artificial Intelligence and Soft Computing (IJAISC) 2023 Vol.8 No.1 pp.62 - 86 Abstract: Dengue is a mosquito-borne arboviral disease infecting humans. Studies analysed the spread of dengue by considering climatic factors alone. However, its transmission does not depend only on climatic factors. This paper focuses on finding the contribution of climatic features in predicting dengue fever and the other factors affecting its transmission. To achieve this, various data preprocessing steps and feature selection are done to drop the non-essential data. Hyper-parameter optimisation is done to tune the different machine learning models, which are then deployed. The best-fit model for San Juan and Iquitos data is based on the evaluation metrics. For San Juan City, the best performance is obtained using the XG Boost model, while using the random forest model for Iquitos City. However, the results explain that using only climatic features is insufficient for predicting dengue cases. Immunity, infecting serotype, hygiene, precautions taken, and the secondary vectors also affect its spread. Inderscience Publishers - linking academia, business and industry through research

Title: Prediction of dengue fever using supervised machine learning based on hyperparameter tuning and an analysis of the factors influencing dengue spread

Authors: Harshita Mandalika; Manya Rampuria; Nenavath Srinivas Naik

Addresses: Department of Electronics and Communication Engineering, International Institute of Information Technology, Naya Raipur, Chattisgarh, India ' College of Arts and Sciences, Georgia State University, Atlanta, USA ' Department of Computer Science and Engineering, Indian Institute of Information Technology, Design and Manufacturing, Kurnool, Andhra Pradesh, India

Abstract: Dengue is a mosquito-borne arboviral disease infecting humans. Studies analysed the spread of dengue by considering climatic factors alone. However, its transmission does not depend only on climatic factors. This paper focuses on finding the contribution of climatic features in predicting dengue fever and the other factors affecting its transmission. To achieve this, various data preprocessing steps and feature selection are done to drop the non-essential data. Hyper-parameter optimisation is done to tune the different machine learning models, which are then deployed. The best-fit model for San Juan and Iquitos data is based on the evaluation metrics. For San Juan City, the best performance is obtained using the XG Boost model, while using the random forest model for Iquitos City. However, the results explain that using only climatic features is insufficient for predicting dengue cases. Immunity, infecting serotype, hygiene, precautions taken, and the secondary vectors also affect its spread.

Keywords: dengue prediction; supervised machine learning; correlation; hyperparameter optimisation; climatic features; factors affecting dengue spread.

DOI: 10.1504/IJAISC.2023.137356

International Journal of Artificial Intelligence and Soft Computing, 2023 Vol.8 No.1, pp.62 - 86

Accepted: 14 Dec 2023
Published online: 13 Mar 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Prediction of dengue fever using supervised machine learning based on hyperparameter tuning and an analysis of the factors influencing dengue spread

Keep up-to-date