Asthma is a chronic and multifactorial respiratory disease that depends on the patient's genetic make-up, the internal environment in the body and the external environment within which the patient exists. These complex influences are explored in this project by employing a large number of patients’ records, 2,392. It consists of numerous demographic characteristics, lifestyles (tobacco use and exercise), environmental conditions (pollution and irritants), past medical history (asthma in first-degree relatives and allergies), physical examination (spirometry), and symptoms (wheezing, breathlessness). Some of the crucial goals of the present work are to determine the factors that would impact the prevalence of Gastroesophageal Reflux and Smoking status among patients with asthma. It involves data cleaning by dealing with the missing values and categorical variables, and feature scaling by using z-score to deal with outliers. The SMOTE algorithm is used on the dataset to cope with the problem of class imbalance. Logistic Regression, Decision Trees, KNN, AdaBoost, and NB algorithms are used to forecast the target variables. There are a lot of evaluation matrices such as accuracy, precision, recall, F1 measure, Receiver Operating Characteristic (ROC) Area Under the Curve, and many more to evaluate the model. ROC curves and precision-recall curves are commonly used to give a full and accurate evaluation of the model’s performance with the help of confusion matrices. The outcomes of the study shall help to increase understanding of factors that underlie asthma and enhance the potential for prognosis and patient management at the individual level