Impute, Select, Decision Tree and Naïve Bayes (ISE-DNC): An ensemble learning approach to classify the Lung Cancer

Abstract

In this work, we have introduced a hybrid novel approach to classify the lung cancer data using ensemble learning. According to this approach, first of all, we present data preprocessing model where missing values are imputed with the help of knn. Later, we incorporated filtering-based feature selection to reduce the feature dimension. Later, decision tree and Naïve Bayes classifiers are used to create the ensemble learner. Finally, voting based decisions are made to classify the data. The proposed approach is represented as ISE-DNC (Impute, Select, Decision Tree and Naïve Bayes) classifier. The proposed approach is implemented on two lung cancer public datasets which are obtained from the UCI repository. The experimental study shows that the proposed approach achieves 96.87% and 89.78% of classification accuracy for lung cancer and Thoracic Surgery Dataset.