Lung cancer kills more people than any other type of cancer, and this is likely to stay true for a long time. Lung cancer can be treated if the signs are found early. If lung cancer symptoms are found early, the latest advances in artificial intelligence can be used to make an experimental diagnosis plan that will work. In this study, optimized support vector machines (SVMs) with an artificial bee colony were used to process the detection from the lung cancer dataset. An SVM classifier is used to categories lung cancer patients based on their symptoms. We examined our ABC-SVM model's balanced accuracy, F1 score, Mathew's correlation coefficient, Sensitivity and specificity to see how well it worked. The evaluated model was trained and tested using benchmark cancer datasets. Irvine. Patients with lung cancer can receive real-time treatment from any location and at any time, with the smallest amount of effort and latency. The suggested model was compared using SVM (linear), SVM (radial basis), GA-based SVM, and PSO-based SVM. When compared to existing methods, the proposed method is 95.65% as accurate.