Lung cancer is one of the most common causes of cancer-related deaths world-wide, and early detection is essential to improving survival rates. There are many difficulties in predicting lung cancer from complicated datasets, such as imaging, genetic, clinical, and environmental data. A new hybrid model called SVM-GBM-GA aims to increase feature selection and prediction accuracy by combining Genetic Algorithms (GA), Gradient Boosting Machine (GBMs) and Support Vector Machine (SVMs). While SVM establishes decision boundaries following feature selection, GA optimizes the model by concentrating on the most predictive variables. GBM improves predictive power by capturing interactions and non-linear relationships in between selected features. With improved predictability and interpretability, the SVM-GBM-GA hybrid model gives clinicians greater understanding of the importance of individual features. Clinical decision-making and the creation of customized treatment plans are aided by this transparency.