Association rule-based feature selection for credit risk assessment

Xueyan Mei, Yilin Jiang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

4 Scopus citations

Abstract

Credit risk assessment becomes rapidly important for credit department of the bank to determine whether to issue credit cards, and make loans to both companies and individualities. However, due to the complexity of database, it is arduous for credit managers to make decisions. In order to solve this problem, this paper proposes a framework that combines feature selection and decision tree classification for credit risk assessment. The feature selection is based on association rules mining. The empirical result shows that our approach can not only reduce the data dimensionality significantly but also provide higher classification accuracies than other methods.

Original languageEnglish
Title of host publicationProceedings of 2016 IEEE International Conference of Online Analysis and Computing Science, ICOACS 2016
EditorsGuorong Chen, Jun Peng
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages301-305
Number of pages5
ISBN (Electronic)9781467377546
DOIs
StatePublished - 17 Sep 2016
Externally publishedYes
Event2016 IEEE International Conference of Online Analysis and Computing Science, ICOACS 2016 - Chongqing, China
Duration: 28 May 201629 May 2016

Publication series

NameProceedings of 2016 IEEE International Conference of Online Analysis and Computing Science, ICOACS 2016

Conference

Conference2016 IEEE International Conference of Online Analysis and Computing Science, ICOACS 2016
Country/TerritoryChina
CityChongqing
Period28/05/1629/05/16

Keywords

  • Association rule mining
  • Credit risk assessment
  • Decision trees
  • Feature Selection

Fingerprint

Dive into the research topics of 'Association rule-based feature selection for credit risk assessment'. Together they form a unique fingerprint.

Cite this