Over the past decade, there has been a marked increase in interest in the automated identification of malignant tumors, largely due to the demand for an early and precise diagnosis that would lead to the best available therapy for the impending risk. As part of this effort, a variety of machine-learning and artificially intelligent approaches have been used to produce reliable aiding tools. To improve the automatic recognition and diagnosis of problematic lung areas, a deep learning model relying on the transfer learning approach is constructed in this research. VGG16, VGG19, and Inception-V3 are employed for the extraction of features from the IQ-OTHNCCD lung cancer dataset. According to experimental findings, transfer-learning models employing the SVM classifier were more effective than those utilizing the softmax function classifier at classifying CT scan images of the used dataset, Results from experiments demonstrate that the VGG16 model is effective for diagnosing lung cancer exceeding other existing models utilizing the same dataset.