Comparing Models and Performance Metrics for Lung Cancer Prediction using Machine Learning Approaches.

Lung cancer is both common and lethal, leading to a significant rise in death rates worldwide. This research focuses on utilizing Machine-Learning (ML) to detect early-stage lung cancer, aiming to address these major public health concerns by using ML to help develop more efficient early detection techniques. It will lower lung cancer rates and improve global healthcare. To achieve these goals, we explored many ML algorithms and compared them using a dataset with lifestyle and health data. The models included Logistic Regression (LR), Random Forest (RF), Naive Bayes (NB), and Support Vector Classifier (SVC). We evaluated these models i.e., based on the evaluation and the key performance metrics. These metrics highlight the benefits and drawbacks of each model. When we compared them, we found that the SVC and LR achieved 84% accuracy. In contrast, NB and RF got 81% accuracy. We performed hyperparameter tuning, which improved LR accuracy to 85%. This enhancement shows that tuning hyperparameters is effective. It optimizes the performance of models for predicting lung cancer.