This study proposes a harmonic average of support and confidence method (HSC) to select important rules from a decision tree, building a core rule-based decision tree (CorDT) that explains insolvency factors in SMEs. An insolvency prediction model was developed using a decision tree algorithm and technological feasibility assessment data, categorized into general, technology development, and toll processing types. Data balancing techniques were applied, with SMOTE showing the highest performance (77.6% hit ratio). Important rules were selected using HSC, and CorDTs were built for each SME type, explaining insolvency causes and suggesting customized prevention strategies.
Publisher
HUMANITIES AND SOCIAL SCIENCES COMMUNICATIONS
Published On
Dec 11, 2023
Authors
Sanghoon Lee, Keunho Choi, Donghee Yoo
Tags
insolvency
decision tree
SME
support and confidence
data balancing
core rule-based decision tree
Related Publications
Explore these studies to deepen your understanding of the subject.
