Dengue Fever Spread Predicting Based on Ensemble Learning and Markov Chain Monte Carlo (MCMC) Method
This study delves into the predictive modeling of Dengue fever transmission, a mosquito-borne disease prevalent in tropical and subtropical regions. Recognizing the potential shifts in disease distribution due to climate change, we employed environmental data collected by various U.S. federal departments to construct a machine learning-based predictive model. After comparing traditional statistical methods with machine learning/deep learning techniques, we opted for a machine learning model for its superior interpretability. To enhance prediction accuracy, we incorporated ensemble learning techniques and combined them with Bayesian methods, specifically the Markov Chain Monte Carlo (MCMC) approach, to ascertain the confidence intervals of our predictions. The model demonstrated commendable performance on the test set, with a Mean Absolute Error (MAE) of 6.39 and a Mean Squared Error (MSE) of 113.50. This research offers an effective methodology for predicting Dengue fever transmission, aiding global public health sectors in better addressing this challenge.
Users could use the code as follows to clone the whole repository.
git clone https://github.com/HironyHan/ELM_prediction elm
This cloned repo is your project install directory. It's helpful to name it something clear and concise like "elm" (as in the final parameter above).