Lung Cancer Prediction using Machine Learning based Feature Selection: A comparative Study

Abstract

One of the disease types that caused more than 2 million fatalities in 2020 was lung cancer. Early diagnosis and treatment of the condition can greatly minimize lung cancer mortality. An early cancer diagnosis is essential for successful treatment and recovery. As a result, the major goal of this research study is to use several machine learning models, including K-Nearest Neighbors, Support Vector Machine, Multilayer Perceptron, and RBF Classifier, to predict the lung cancer disease from various symptoms. A text dataset with 15 features is used to assess the ML algorithms. The algorithms are compared based on how well the algorithms perform when using different feature selection and extraction techniques. The models' performance evaluation shows that MLP is achieving a consistent result with and without feature selection methods with an accuracy around 90%.