Start Submission Become a Reviewer

Reading: A Data-Driven Method for Selecting Optimal Models Based on Graphical Visualisation of Differ...

Download

A- A+
dyslexia friendly

Proceedings Papers

A Data-Driven Method for Selecting Optimal Models Based on Graphical Visualisation of Differences in Sequentially Fitted ROC Model Parameters

Authors:

K S Mwitondi ,

Sheffield Hallam University, Faculty of Arts, Computing, Engineering and Sciences, Sheffield S1 1WB, UK, US
X close

R E Moustafa,

George Washington University, Statistics Department, 2140 Pennsylvania Ave., NW, Washington DC, 20052, USA, US
X close

A S Hadi

The American University in Cairo, Egypt/Cornell University, 291 Ives Hall, Cornell University, Ithaca, NY 14853-3901, USA, US
X close

Abstract

Differences in modelling techniques and model performance assessments typically impinge on the quality of knowledge extraction from data. We propose an algorithm for determining optimal patterns in data by separately training and testing three decision tree models in the Pima Indians Diabetes and the Bupa Liver Disorders datasets. Model performance is assessed using ROC curves and the Youden Index. Moving differences between sequential fitted parameters are then extracted, and their respective probability density estimations are used to track their variability using an iterative graphical data visualisation technique developed for this purpose. Our results show that the proposed strategy separates the groups more robustly than the plain ROC/Youden approach, eliminates obscurity, and minimizes over-fitting. Further, the algorithm can easily be understood by non-specialists and demonstrates multi-disciplinary compliance.

DOI: http://doi.org/10.2481/dsj.WDS-045
How to Cite: Mwitondi, K.S., Moustafa, R.E. & Hadi, A.S., (2013). A Data-Driven Method for Selecting Optimal Models Based on Graphical Visualisation of Differences in Sequentially Fitted ROC Model Parameters. Data Science Journal. 12, pp.WDS247–WDS253. DOI: http://doi.org/10.2481/dsj.WDS-045
4
Downloads
1
Citations
Published on 02 May 2013.
Peer Reviewed

Downloads

  • PDF (EN)

    comments powered by Disqus