Comparative analysis of data mining methods for bankruptcy prediction

David L. Olson, Dursun Delen, Yanyan Meng

Research output: Contribution to journalArticle

115 Citations (Scopus)

Abstract

A great deal of research has been devoted to prediction of bankruptcy, to include application of data mining. Neural networks, support vector machines, and other algorithms often fit data well, but because of lack of comprehensibility, they are considered black box technologies. Conversely, decision trees are more comprehensible by human users. However, sometimes far too many rules result in another form of incomprehensibility. The number of rules obtained from decision tree algorithms can be controlled to some degree through setting different minimum support levels. This study applies a variety of data mining tools to bankruptcy data, with the purpose of comparing accuracy and number of rules. For this data, decision trees were found to be relatively more accurate compared to neural networks and support vector machines, but there were more rule nodes than desired. Adjustment of minimum support yielded more tractable rule sets.

Original languageEnglish
Pages (from-to)464-473
Number of pages10
JournalDecision Support Systems
Volume52
Issue number2
DOIs
StatePublished - 1 Jan 2012

Fingerprint

Bankruptcy
Decision Trees
Data Mining
Decision trees
Data mining
Support vector machines
Neural networks
Technology
Research
Bankruptcy prediction
Comparative analysis
Decision tree
Prediction
Comparative Analysis
Decision Tree
Support Vector Machine
Support vector machine
Neural Networks

Keywords

  • Bankruptcy prediction
  • Data mining
  • Decision trees
  • Neural networks
  • Support vector machines
  • Transparency
  • Transportability

Cite this

Olson, David L. ; Delen, Dursun ; Meng, Yanyan. / Comparative analysis of data mining methods for bankruptcy prediction. In: Decision Support Systems. 2012 ; Vol. 52, No. 2. pp. 464-473.
@article{c09a025114a34b978cba2bf58a6fe47f,
title = "Comparative analysis of data mining methods for bankruptcy prediction",
abstract = "A great deal of research has been devoted to prediction of bankruptcy, to include application of data mining. Neural networks, support vector machines, and other algorithms often fit data well, but because of lack of comprehensibility, they are considered black box technologies. Conversely, decision trees are more comprehensible by human users. However, sometimes far too many rules result in another form of incomprehensibility. The number of rules obtained from decision tree algorithms can be controlled to some degree through setting different minimum support levels. This study applies a variety of data mining tools to bankruptcy data, with the purpose of comparing accuracy and number of rules. For this data, decision trees were found to be relatively more accurate compared to neural networks and support vector machines, but there were more rule nodes than desired. Adjustment of minimum support yielded more tractable rule sets.",
keywords = "Bankruptcy prediction, Data mining, Decision trees, Neural networks, Support vector machines, Transparency, Transportability",
author = "Olson, {David L.} and Dursun Delen and Yanyan Meng",
year = "2012",
month = "1",
day = "1",
doi = "10.1016/j.dss.2011.10.007",
language = "English",
volume = "52",
pages = "464--473",
journal = "Decision Support Systems",
issn = "0167-9236",
publisher = "Elsevier",
number = "2",

}

Comparative analysis of data mining methods for bankruptcy prediction. / Olson, David L.; Delen, Dursun; Meng, Yanyan.

In: Decision Support Systems, Vol. 52, No. 2, 01.01.2012, p. 464-473.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Comparative analysis of data mining methods for bankruptcy prediction

AU - Olson, David L.

AU - Delen, Dursun

AU - Meng, Yanyan

PY - 2012/1/1

Y1 - 2012/1/1

N2 - A great deal of research has been devoted to prediction of bankruptcy, to include application of data mining. Neural networks, support vector machines, and other algorithms often fit data well, but because of lack of comprehensibility, they are considered black box technologies. Conversely, decision trees are more comprehensible by human users. However, sometimes far too many rules result in another form of incomprehensibility. The number of rules obtained from decision tree algorithms can be controlled to some degree through setting different minimum support levels. This study applies a variety of data mining tools to bankruptcy data, with the purpose of comparing accuracy and number of rules. For this data, decision trees were found to be relatively more accurate compared to neural networks and support vector machines, but there were more rule nodes than desired. Adjustment of minimum support yielded more tractable rule sets.

AB - A great deal of research has been devoted to prediction of bankruptcy, to include application of data mining. Neural networks, support vector machines, and other algorithms often fit data well, but because of lack of comprehensibility, they are considered black box technologies. Conversely, decision trees are more comprehensible by human users. However, sometimes far too many rules result in another form of incomprehensibility. The number of rules obtained from decision tree algorithms can be controlled to some degree through setting different minimum support levels. This study applies a variety of data mining tools to bankruptcy data, with the purpose of comparing accuracy and number of rules. For this data, decision trees were found to be relatively more accurate compared to neural networks and support vector machines, but there were more rule nodes than desired. Adjustment of minimum support yielded more tractable rule sets.

KW - Bankruptcy prediction

KW - Data mining

KW - Decision trees

KW - Neural networks

KW - Support vector machines

KW - Transparency

KW - Transportability

UR - http://www.scopus.com/inward/record.url?scp=82255192254&partnerID=8YFLogxK

U2 - 10.1016/j.dss.2011.10.007

DO - 10.1016/j.dss.2011.10.007

M3 - Article

AN - SCOPUS:82255192254

VL - 52

SP - 464

EP - 473

JO - Decision Support Systems

JF - Decision Support Systems

SN - 0167-9236

IS - 2

ER -