Lung cancer is a global health concern that has invoked intense interest in predictive modeling that uses machine learning to forecast lung cancer signs and diagnosis. A machine learning, deep learning, feature engineering, and assembly method-based lung cancer prediction system is developed using available data. At the same time, the study does a thorough evaluation of algorithm performance using several feature selection techniques and hyperparameter tuning setups on two different datasets: LCD_1 and LCD_2. With a remarkable accuracy of 99.67%, support vector machine (SVM) is the best-performing algorithm in LCD_1, but logistic regression consistently shows excellent accuracy. Logistic regression regularly produced excellent accuracy in LCD_2, with decision tree achieving the second accuracy.