Analysis of Comparative Methods for Machine Learning-Based Lung Cancer Prediction

The older tiers and led expert documents are used to detect cancer; this could be useful. There has to be room for human mistake. It results in the possibility of inaccuracy in the identification of lung cancer; hence an automatic process is required. In order to reduce human error and improve the accuracy and error-free nature of the system, the report then aims to detect lung cancer early through an automated procedure. To achieve this, machine learning algorithms like naive bayes, KNN, Ridge Classifier, Multilayered Perception, and support vector machine algorithms are used to detect lung cancer. Using a dataset of lung cancer cases, we have developed five algorithms to assess performance according to four parameters: accuracy, recall, f1 score, and precision. Comparative research has been conducted on these parameters, and an AUC-ROC curve has been created to demonstrate how well machine learning models perform.