Chronic kidney disease (CKD) is a widespread illness affecting humans globally. It poses a significant challenge for societies and global health care systems. Specialized screening, diagnosis, and treatment by clinicians are needed to prevent complications due to CKD, and early detection of people with CKD is essential in preventing the progression of the disease. In this study, we present an approach to detect CKD effectively. The approach involves using the information gain (IG) technique to obtain the most relevant CKD features from the dataset. The effectiveness of the feature selection technique was evaluated using logistic regression (LR), decision tree, and support vector machine (SVM) algorithms. Furthermore, these machine learning algorithms were applied individually as base learners in building robust boosted classifiers using the adaptive boosting (AdaBoost) technique to enhance their classification performance. The simulation results indicate that using the IG based feature selection improved the performance of the classifiers, and the boosted classifiers obtained better performance than their corresponding standard versions. Meanwhile, the boosted decision tree (AdaBoost-DT) got the most outstanding performance with accuracy, precision, sensitivity, and F-measure values of 1.000.