Lung Cancer Prediction Using Random Forest

Background: In recent years, lung cancer is a common cancer across the globe. For the early prediction of lung cancer, medical practitioners and researchers require an efficient predictive model, which will reduce the number of deaths. This paper proposes a lung cancer prediction model by using the Random Forest Classifier, which aims at analyzing symptoms (gender, age, air pollution, weight loss, etc.). Objective: This work addresses the problem of classification of lung cancer data using the Random Forest Algorithm. Random Forest is the most accurate learning algorithm and many researchers in the healthcare domain use it. Methods: This paper deals with the prediction of lung cancer by using the Random Forest Classifier. Results: The proposed method (Random Forest Classifier) applied on two lung cancer datasets achieved an accuracy of 100% for the lung cancer dataset-1 and 96.31 on dataset-2. In the prediction of lung cancer, the Random Forest Algorithm showed improved accuracy compared with other methods. Conclusion: This predictive model will help health professionals in predicting lung cancer at an early stage.