Proposal for Measuring Quality of Decision Trees Partition

Proposal for Measuring Quality of Decision Trees Partition

Souad Taleb Zouggar (University of Oran 2, Department of Economics, Oran, Algeria) and Abdelkader Adla (University of Oran 1, Department of Computer Science, Oran, Algeria)
Copyright: © 2017 |Pages: 21
DOI: 10.4018/IJDSST.2017100102
OnDemand PDF Download:
No Current Special Offers


To compute a partition quality for a decision tree, we propose a new measure called NIM “New Information Measure”. The measure is simpler, provides similar performance, and sometimes outperforms the existing measures used with tree-based methods. The experimental results using the MONITDIAB application (Taleb & Atmani, 2013) and datasets from the UCI repository (Asuncion & Newman, 2007) confirm the classification capabilities of our proposal in comparison to the Shannon measure used with ID3 and C4.5 decision tree methods.
Article Preview


Decision Trees

Decision trees methods are widely used in the classification field; they provide good results as far as classification techniques are concerned, but involve complexity in the formulas used for calculations. They enable building classification models as the latter closely resemble human reasoning and are easy to understand. To have more information about decision trees methods, the reader can refer to (Kotsiantis, 2013); the paper describes basic decision trees issues and current research points. In the chapter published by (Barros, 2015) are presented in details the most common approaches for decision-tree induction (Top Down Induction) with brief comments on some alternative construction methods summarizing the main design options to building decision-tree induction algorithms.

Complete Article List

Search this Journal:
Open Access Articles
Volume 14: 4 Issues (2022): Forthcoming, Available for Pre-Order
Volume 13: 4 Issues (2021): 3 Released, 1 Forthcoming
Volume 12: 4 Issues (2020)
Volume 11: 4 Issues (2019)
Volume 10: 4 Issues (2018)
Volume 9: 4 Issues (2017)
Volume 8: 4 Issues (2016)
Volume 7: 4 Issues (2015)
Volume 6: 4 Issues (2014)
Volume 5: 4 Issues (2013)
Volume 4: 4 Issues (2012)
Volume 3: 4 Issues (2011)
Volume 2: 4 Issues (2010)
Volume 1: 4 Issues (2009)
View Complete Journal Contents Listing