Lung Cancer Detection Using Feature Selection and Navie Bayes Classification

We demonstrate NB and Feature Selection, two highly important algorithms for completing this task due to their simplicity and great performance. The classification accuracy of NB and Feature Selection's class chance estimations on the training data is weighted in NB-Feature Selection. The NB-Feature Selection hybrid classification method, which is presented in this research, greatly beats NB and feature Selection in regards to accuracy in classification. It is built on top of a Bayesian network, which consists of a structural model and a number of conditional probabilities. Data mining has a number of key problems and knowledge discovery is classification, and the objective is to develop a classifier from a set of cases with class labels. Many learning algorithms, including decision trees and Bayesian networks, have been created. There are two groups into which these algorithms can be split: probabilitybased algorithms and decision boundary-based algorithms. The decision tree is the most popular classification model, and it categorises a test instance by sorting it through the tree from its root node to one of its leaves, and does so by simply voting.