Abstract:
This research is about finding better ways to predict and manage dengue outbreaks, a
serious illness spread by mosquitoes. To enhance predictions, we are utilizing sophisticated
machine learning methods and intelligent computer programs. Put simply, our focus is on
understanding and forecasting potential dengue outbreak sites and times. The study
highlights the significance of trustworthy data, moral data management, and machine
learning proficiency. Using a dataset that includes important variables like year, month,
area, affected population, and total population, our goal is to create a tool that can quickly
assess risk for people traveling to new places. Based on input parameters like area and
month, we hope to use advanced algorithms to classify the likelihood of dengue occurrence
into four categories: Low, Medium, High, and Risky. Input output analysis prioritizes
privacy protection while ensuring data organization and utility. We prioritize privacy while
making sure the data is well organized and helpful for our computer programs. Our project
management guarantees an organized strategy with efficient teamwork, financial analysis,
budgeting, and flexibility in the face of difficulties. In summary, the goal of this research
is to provide important new information for the comprehension and forecasting of dengue
outbreaks. The methodical approach, which is based on state-of-the-art machine learning
techniques and data-centric approaches, aims to close current gaps and offer a strong
framework for managing and predicting dengue outbreaks. After a thorough analysis of the
data, the Decision Tree algorithm was the best performer, with an astounding 84%
accuracy, 60% F1 score, 67% precision, and 55% recall rate. Considering these results, the
Decision Tree is the recommended algorithm for dengue epidemic prediction.
Consequently, decision trees are the best option for this purpose since we can accurately
and consistently predict dengue outbreaks.
