The diagnosis and detection of lung cancer are critical due to its high mortality rates and the challenges posed by its unknown causes. Early detection is paramount for effective treatment, as lung carcinoma often progresses rapidly. While prevention is difficult, reducing risk factors such as smoking can help mitigate the disease's impact. Detecting lung cancer in its earliest stages significantly improves patient survival rates. Predictive models have been developed to aid clinicians in managing indeterminate lung nodules, potentially reducing unnecessary follow-ups for benign nodules. This study focuses on the early diagnosis of lung cancer, evaluating the performance of classification algorithms like Naive Bayes, SVM, Decision Trees, and Logistic Regression. By analysing these algorithms' efficacy, the aim is to enhance lung cancer detection and ultimately improve patient outcomes.