Meteorological factors cannot be ignored in machine learning-based methods for predicting dengue, a systematic...

Area of Science:

• Epidemiology
• Data Science
• Public Health

Background:

• Dengue fever incidence prediction is increasingly utilizing machine learning.
• Significant variability exists in predictive factors and models across studies.
• A systematic review is needed to consolidate current knowledge.

Approach:

• Conducted a systematic review of studies published up to July 2023.
• Searched PubMed, ScienceDirect, and Web of Science databases.
• Included 23 papers focusing on dengue incidence forecasting and machine learning methods.

Key Points:

• Meteorological factors (temperature, rainfall, humidity) are crucial predictors.
• Historical dengue data and environmental factors are also significant.
• Support Vector Machine (SVM) and Long Short-Term Memory (LSTM) demonstrate strong predictive performance.

Conclusions:

• Meteorological factors are essential for accurate dengue fever prediction.
• SVM and LSTM algorithms are the most effective machine learning models identified.
• This review supports enhanced dengue early warning systems and infectious disease prediction.