Asthma is a chronic respiratory condition that affects millions of individuals worldwide, requiring prompt attention and accurate prediction for necessary treatments. This research integrates fuzzification and machine learning techniques to build an ensemble classifier for asthma prediction. Fuzzification handles the uncertainty and imprecision inherently present in medical data by transforming numerical attributes into fuzzy sets using appropriate membership functions. Machine learning models, including Decision Tree, Logistic Regression, and Neural Network algorithms, are utilized to build individual classifiers. To improve classification accuracy, the bagging technique is employed to combine predictions from these classifiers into an ensemble model. The implementation is carried out using Python for fuzzification and the Orange tool for machine learning. The proposed solution demonstrates the potential of combining fuzzification with machine learning techniques to predict asthma more accurately compared to conventional approaches. Tree, Naive Bayes, Logistic Regression, kNN, SVM, Neural Network and Random Forest methods are evaluated. Additionally, we used the bagging technique to build an ensemble classifier, improving accuracy by providing output based on analyzing the outputs of individual classifiers. Compared to the initial performance analysis, accuracy improved after feature selection, as it identified the most relevant features for building the model. The accuracy was further improved after fuzzification, as it handled the possibility of uncertainty by transforming numerical features into categorical ones.