Lung Cancer Prediction System based on Machine Learning Algorithms

Abstract

Lung cancer is the most dangerous malignant tumour in terms of morbidity and mortality, and it poses a significant threat to human health. Recognizing and predicting lung cancer at the earliest possible stage can significantly enhance patient survival. Machine learning techniques can predict lung cancer early and effectively. We used a publicly accessible dataset from the Kaggle web repository and employed machine learning algorithms to predict lung cancer. After the pre- processing and normalization procedures on the dataset, the dataset is divided into training and testing subsets. To determine the optimum model for lung cancer prediction, this study employs four prominent classifiers, such as AdaBoost, Decision Tree (DT), Support Vector Machine (SVM), and Random Forest (RF). In this study, experimental results show that the proposed machine learning algorithms achieve accuracy of 71.67%, 85.67%, 97%, and 100% in predicting three levels (low, medium, and high) of lung cancer. Random Forest classifier outperforms the other classifiers with the highest accuracy. The performance of classifiers is compared using parameters such as precision, recall, F1 score, and accuracy.