UPM Institutional Repository

Using feature selection as accuracy benchmarking in clinical data mining.


Hossain, Jafreen and Mohd. Sani, Nor Fazlida and Mustapha, Aida and Affendey, Lilly Suriani (2013) Using feature selection as accuracy benchmarking in clinical data mining. Journal of Computer Science, 9 (7). pp. 883-888. ISSN 1549-3636


Automated prediction of new patients’ disease diagnosis based on data mining analysis on historical data is proven to be an extremely useful tool in the medical innovation. There are several studies focusing on this particular aspect. The objective of this study is two-fold. First, we look into three different classifiers, which are the Naïve Bayes, Multilayer Perceptron (MLP) and Decision Tree J48 to predict the diagnosis results. Next, we investigate the effects of feature selection in such experiments. We also compare the experimental results with the study of Comparative Disease Profile (CDP) using the same dataset. Results have shown that the Naive Bayes provides the best result in terms of accuracy in our experiments and in comparison with CDP. However, we suggest using Multilayer Perceptron since the variables used in our experiments are inter-dependent among each other. In addition, MLP has shown better accuracy than CDP.

Download File

PDF (Abstract)
Using feature selection as accuracy benchmarking in clinical data mining.pdf

Download (182kB) | Preview
Official URL or Download Paper: http://thescipub.com/issue-jcs/9/7

Additional Metadata

Item Type: Article
Divisions: Faculty of Computer Science and Information Technology
DOI Number: https://doi.org/10.3844/jcssp.2013.883.888
Publisher: Science Publications
Keywords: Data mining; Healthcare; Heart disease; Multilayer perceptron; Naive bayes; J48.
Depositing User: Ms. Nida Hidayati Ghazali
Date Deposited: 03 Jun 2014 08:18
Last Modified: 11 Sep 2015 03:48
Altmetrics: http://www.altmetric.com/details.php?domain=psasir.upm.edu.my&doi=10.3844/jcssp.2013.883.888
URI: http://psasir.upm.edu.my/id/eprint/30669
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item