Hyperparameter Tuning In Ensemble Learning Models

Abstract

Effective detection systems, for diseases play a role in healthcare helping with diagnosis and treatment. This study offers a comparison of hyper parameter tuning methods for disease detection systems using four healthcare datasets; kidney disease, diabetes detection, heart disease and breast cancer detection. The main goals of this research involve preparing the datasets by standardizing the entries and testing machine learning models such as Support Vector Machines (SVM) Naive Bayes, Logistic Regression and k Nearest Neighbors (kNN) to determine the effective model for each dataset. After implementing the models we apply three hyper parameter tuning techniques; Grid Search, Random Search and Particle Swarm Optimization (PSO). These methods are used to tune the model parameters. Enhance overall performance metrics. The evaluation focuses on accuracy metrics to compare how the models perform before and after hyper parameter tuning. The outcomes of this study illustrate how different tuning techniques can enhance the performance of disease detection systems across a range of healthcare datasets. By conducting experiments and analysis we pinpoint the suitable tuning method for each dataset offering valuable insights, for developing precise and efficient disease detection systems. These discoveries play a role, in pushing forward the realm of healthcare analytics and machine learning leading to patient results and healthcare services.