Improving the efficiency of lung cancer data classification using ensemble technique

Abstract

The objectives of this research are 1) to compare the efficiency of lung cancer data classification using four data mining techniques: Decision Tree, Naive Bayes, Support Vector Machine, and Artificial Neural Network; 2) to improve the data classification model for lung cancer datasets which consisted of 16 columns and 310 rows. The performance was tested by the 10-Fold Cross Validation method using Rapid Miner Studio 9.10 software. The results show that the Artificial Neural Network gave the best efficiency with an accuracy of 89.25 %, recall of 90.12 %, and precision of 97.33 %. After that, the efficiency of data classification of the Artificial Neural Network is improved by using the majority vote ensemble method. It improved the efficiency with a classification accuracy of 91.40 % and a recall of 92.59 % and a precision is 97.40 %, respectively.