A data mining based inverse classification with hybrid genetic approaches

Mu-Chen Chen; Chia Ping Lin; Shih Hsien Huang; Wei Rong Zeng

A data mining based inverse classification with hybrid genetic approaches

Mu-Chen Chen^*, Chia Ping Lin, Shih Hsien Huang, Wei Rong Zeng

^*Corresponding author for this work

National Chiao Tung University

Research output: Contribution to conference › Paper › peer-review

Abstract

The credit scoring models are developed to categorize applicants as either accept or reject with respect to their characteristics, and thereby to minimize the creditors' risk and translate considerably into future savings. In this paper, we use Genetic Programming (GP) as a classification system to build the credit scoring model. The post classification analysis, namely inverse classification, is adopted to better understand rejected credits, and try to reassign them to the accept class. An optimization method based on genetic algorithm (GA) is used to reassign the rejected instances to the accept class for balancing between adjustment cost and customer preference.

Original language	English
Pages	1668-1675
Number of pages	8
State	Published - 1 Dec 2006
Event	36th International Conference on Computers and Industrial Engineering, ICC and IE 2006 - Taipei, Taiwan Duration: 20 Jun 2006 → 23 Jun 2006

Conference

Conference	36th International Conference on Computers and Industrial Engineering, ICC and IE 2006
Country	Taiwan
City	Taipei
Period	20/06/06 → 23/06/06

Keywords

Credit scoring
Genetic algorithm
Genetic programming
Inverse classification

Fingerprint Dive into the research topics of 'A data mining based inverse classification with hybrid genetic approaches'. Together they form a unique fingerprint.

Cite this

@conference{a5c50167a2b9483ebb7167f3f623fac0,

title = "A data mining based inverse classification with hybrid genetic approaches",

abstract = "The credit scoring models are developed to categorize applicants as either accept or reject with respect to their characteristics, and thereby to minimize the creditors' risk and translate considerably into future savings. In this paper, we use Genetic Programming (GP) as a classification system to build the credit scoring model. The post classification analysis, namely inverse classification, is adopted to better understand rejected credits, and try to reassign them to the accept class. An optimization method based on genetic algorithm (GA) is used to reassign the rejected instances to the accept class for balancing between adjustment cost and customer preference.",

keywords = "Credit scoring, Genetic algorithm, Genetic programming, Inverse classification",

author = "Mu-Chen Chen and Lin, {Chia Ping} and Huang, {Shih Hsien} and Zeng, {Wei Rong}",

year = "2006",

month = dec,

day = "1",

language = "English",

pages = "1668--1675",

note = "null ; Conference date: 20-06-2006 Through 23-06-2006",

}

TY - CONF

T1 - A data mining based inverse classification with hybrid genetic approaches

AU - Chen, Mu-Chen

AU - Lin, Chia Ping

AU - Huang, Shih Hsien

AU - Zeng, Wei Rong

PY - 2006/12/1

Y1 - 2006/12/1

N2 - The credit scoring models are developed to categorize applicants as either accept or reject with respect to their characteristics, and thereby to minimize the creditors' risk and translate considerably into future savings. In this paper, we use Genetic Programming (GP) as a classification system to build the credit scoring model. The post classification analysis, namely inverse classification, is adopted to better understand rejected credits, and try to reassign them to the accept class. An optimization method based on genetic algorithm (GA) is used to reassign the rejected instances to the accept class for balancing between adjustment cost and customer preference.

AB - The credit scoring models are developed to categorize applicants as either accept or reject with respect to their characteristics, and thereby to minimize the creditors' risk and translate considerably into future savings. In this paper, we use Genetic Programming (GP) as a classification system to build the credit scoring model. The post classification analysis, namely inverse classification, is adopted to better understand rejected credits, and try to reassign them to the accept class. An optimization method based on genetic algorithm (GA) is used to reassign the rejected instances to the accept class for balancing between adjustment cost and customer preference.

KW - Credit scoring

KW - Genetic algorithm

KW - Genetic programming

KW - Inverse classification

UR - http://www.scopus.com/inward/record.url?scp=84886914153&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:84886914153

SP - 1668

EP - 1675

Y2 - 20 June 2006 through 23 June 2006

ER -

A data mining based inverse classification with hybrid genetic approaches

Abstract

Conference

Keywords

Other files and links

Cite this