Artificial Intelligence (AI) and Machine learning (ML) models have proven to be scalable approaches for handling several biomedical problems. Recent availability of high-quality datasets which captures various factors contributing to the respiratory disease, have enabled the development of robust models that deliver high accuracy and precision scores in early detection of respiratory diseases. This paper focuses on asthma disease detection. It makes two primary contributions: (1) an empirical evaluation of their performance on an asthma disease detection dataset with data mining and pre-processing techniques, and (2) the identification of the most effective approach for asthma disease detection based on rigorous evaluation using metrics such as precision, accuracy, recall, F-1, and F-beta scores. Cat Boost Classifier was found to be the best model which predicted asthma disease with a 96.04% accuracy.