<![CDATA[Vector-borne diseases such as malaria, dengue fever and yellow fever still pose a serious threat to public health. To distinguish between these diseases, an accurate classification process is required. In this study, Random Forest algorithm is used as a classification method due to its ability to overcome overfitting and provide good accuracy results. However, the large number of features in the data can cause redundancy and decrease the accuracy of the model. Therefore, the Binary Particle Swarm Optimization (BPSO) method is used as a feature selection technique to optimize the performance of Random Forest. The optimization process is also complemented by finding the best parameters using Random Search and Grid Search. Evaluation was conducted on a vector-borne disease dataset with 64 features and 11 disease classes. The results showed that the accuracy of the model increased from 90.48% to 100% after feature selection by BPSO which selected 37 best features, and Random Search proved to be more efficient in computation time than Grid Search. This research shows that the combination of Random Forest and BPSO can improve classification accuracy and efficiency in detecting vector-borne diseases.]]>
Copyrights © 2025
