Abstract
This study is to compare the prediction accuracy of multilayer perceptron in neural networks against tree-based algorithms, in particular the ID3 and J48 algorithms on Pima Indian diabetes mellitus data set. The classification experiment is performed using algorithms in WEKA to determine the class diabetes or non-diabetes with the data set of 768 patients. Results showed that a pruned J48 tree performed with higher accuracy, which is 89.3% as compared to 81.9% by the multilayer perceptrons. On further removal of the number of times pregnant attribute, the prediction accuracy for the pruned J48 tree improved to 89.7%.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Malaysian Diabetes Association, (2010), http://www.diabetes.org.my/article.php?aid=5 (retrieved from)
Piatetsky–Shapiro, G.: Knowledge Discovery in Real Databases: A Report on the IJCAI-89 Workshop. AI Magazine 11(5), 68–70 (1991)
Jaafar, S.F., Ali, D.M.: Diabetes Mellitus Forecast using Artificial Neural Network (2005)
Han, J., Rodriguez, J.C., Beheshti, M.: Discovering Decision Tree-based Diabetes Prediction Model. Department of Computer Science. California State University (2009)
Andrei, D., Viviana, A.: Overview on How Data Mining Tools may Support Cardiovascular Disease Prediction. Journal of Applied Computer Science & Mathematics 8(4), 57–62 (2010)
Fisher, D.H., McKusick, K.B.: An Empirical Comparison of ID3 and Backpropagation. In: Eleventh International Joint Conference on Artificia1 Intelligence, Detroit, MI, August 20-25, pp. 788–793 (1989)
Mooney, R., Shalvik, J., Towell, G.: Symbolic and Neural Learning Algorithms: An Experimental Comparison. Machine Learning 6, 11–143 (1991)
Mooney, R., Shavlik, J., Towell, G.,, G.: An Experimental Comparison of Symbolic and Connectionist Learning Algorithms. In: Eleventh International Joint Conference on Artificial Intelligence, Detroit, MI, August 20-25, pp. 77–780 (1989)
Konda, S.R.: A Comparative Evaluation of Symbolic Learning Methods and Neural Learning Methods
Chen, J., Luo, D.L., Mu, F.X.: An Improved ID3 Decision Tree Algorithm. In: Fourth International Conference on Computer Science and Education (2009)
C4.5 Algorithm (2010), http://en.wikipedia.org/wiki/C4.5_algorithm (retrieved from)
Xiaoliang, Z., Jian, W., Hongcan, Y., Shangzhuo, W.: Research and Application of the Improved Algorithm C4.5 on Decision Tree. In: International Conference on Test and Measurement (2009)
Hall, M., Frank, E., Holmes, G., Pfahringer, B., Reutemann, P., Witten, I.H.: The WEKA Data Mining Software: An Update; SIGKDD Explorations, vol. 11(1) (2009)
Asuncion, A., Newman, D.J.: UCI Machine Learning Repository. University of California, Department of Information and Computer Science, Irvine, (2010), http://www.ics.uci.edu/~mlearn/MLRepository.html (retrieved from)
Karegowda, A.G., Manjunath, A.S., Jayaram, M.A.: Comparative Study of Attribute Selection using Gain Ratio and Correlation-based Feature Selection. International Journal of Information Technology and Knowledge Management 2(2), 271–277 (2010)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Ahmad, A., Mustapha, A., Zahadi, E.D., Masah, N., Yahaya, N.Y. (2011). Comparison between Neural Networks against Decision Tree in Improving Prediction Accuracy for Diabetes Mellitus. In: Snasel, V., Platos, J., El-Qawasmeh, E. (eds) Digital Information Processing and Communications. ICDIPC 2011. Communications in Computer and Information Science, vol 188. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-22389-1_47
Download citation
DOI: https://doi.org/10.1007/978-3-642-22389-1_47
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-22388-4
Online ISBN: 978-3-642-22389-1
eBook Packages: Computer ScienceComputer Science (R0)