Performance Analysis of Different Machine Learning Classifiers for Prediction of Lung Cancer

Cancer is, beyond doubt, among the most significant causes of death today. Cancer continues to be a major mortality factor despite several decades of clinical research and experiments of new treatments. It can occur in any part of the body, including the lungs. Primary lung cancer symptoms frequently lack specificity and could be linked to smoking. In clinical and medical data analysis, the prediction of lung cancer is a difficult task. A subdivision of artificial intelligence, also called “machine learning,” employs distinguished analytical, stochastic, and optimization techniques for helping machines to be trained from past understandings and analyze extensive and diverse data sets. As a result, machine learning is widely utilized in the treatment and prediction of cancer. Machine learning (ML) classifiers are useful in contributing to the making of decisions and forecasting the severity of cancer by using cosmic amounts of data. Through the mediums of this study, we have proposed some classification algorithms to deter the existence of lung cancer in a person’s body influenced by the symptoms one experiences. Different machine language classifiers are implemented over the Lung cancer dataset. With 93% precision, the accuracy of the SVM classifier has been the highest. A new ensembled model has been introduced with the help of ensemble learning which combines three different models – Logistic Regression (LR), KNN and Random Forest (RF). The accuracy achieved using applied ensemble model is 93.5%.