An Empirical Analysis of Classification Trees Algorithm for Protein Datsets

R. Ranjani Rani, P. Manikandan, Dr.D. Ramya Chitra


In general, the Classification tree can be used to predict membership of cases or objects in the classes of a categorical dependent variable from their measurements on one or more predictor variables. Generally, Classification tree analysis is one of the major techniques used in so-called Data Mining. In this paper we are analyzing the performance of 4 classifiers trees algorithms namely J48 Decision tree, Naïve Bayes Tree, Random Forest and Random tree. In this article we used protein datasets namely the dengue virus and the Superoxide Dismutase1 (SOD1) protein datasets for calculating the performance by using the cross validation parameter. And finally we performed the comparative analysis based on the factors such as the classification accuracy, performance and error rate measures on all the algorithms.


Keywords: Data Mining, Classification, J48 Decision Tree, Random Forest, Random Tree, NaïveBayes Tree, Dengue, SOD1 datasets.

Full Text:




  • There are currently no refbacks.

Copyright (c) 2016 International Journal of Advanced Research in Computer Science