|
Title:
|
SOCIAL NETWORKS IN CREDIT SCORING: A MACHINE LEARNING APPROACH |
|
Author(s):
|
Ahmad Abd Rabuh, Mark Xu and Renatas Kizys |
|
ISBN:
|
978-989-8704-50-4 |
|
Editors:
|
Piet Kommers, Mário Macedo, Guo Chao Peng and Ajith Abraham |
|
Year:
|
2023 |
|
Edition:
|
Single |
|
Keywords:
|
Machine Learning, Predictive Analytics, Social Networks, Credit Scoring, Financial Inclusion |
|
Type:
|
Full |
|
First Page:
|
277 |
|
Last Page:
|
283 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
This research examines if social network tie has an incremental predictive ability for borrower default in credit scoring and the precision of effect. With advanced digital technologies and increasing availability of non-financial behaviour big data, social network data has been explored to assess consumer credit scoring in research and practice. This research uses machine learning algorithms to analyse a large dataset (loan applications and defaults) obtained from a European lender. The results show that social network data, when working together with traditional financial data, improves predictive ability of borrowers' default. A Bayesian Analysis confirms the explanatory evidence of social ties of borrowers. This research generates insights of machine learning power in analyzing imbalanced large dataset using ten different classifiers, and contributes to the theoretical debate on social capital theory as well as practical guidance of using XGBoost algorithm for lenders. |
|
|
|
|
|
|
