Feature selection in peer-to-peer lending based on hybrid modified grey wolf optimization with optimized decision tree for credit risk assessment

ABSTRACT

Lending platforms operating on a peer-to-peer (P2P) basis encounter the intricate challenge of assessing borrower creditworthiness to minimize the risk of defaults. This study addresses this challenge by proposing an advanced approach to feature selection that leverages the Grey Wolf Optimizer (GWO) in conjunction with a finely tuned Decision Tree (DT) model. The main objective is to enhance the precision and efficiency of feature selection processes within P2P lending datasets. The study begins by fine-tuning DT hyperparameters using Genetic Algorithms (GA), yielding an optimal configuration: ‘max_depth’ = 40, ‘min_samples_leaf’ = 20, and ‘criterion’ = ‘entropy’. Subsequent phases involve the application of GWO and modified GWO (nGWO, cGWO, and lGWO) to conduct feature selection under distinct Search Agent (SA) setups (SA = 5, SA = 20, SA = 50). Particularly impressive is the performance of the lGWO model with the SA = 50 setup, achieving a remarkable 91% accuracy while selecting 80.55% of the total 36 features. This study significantly improves how lenders manage risks in P2P lending by identifying high-risk borrowers more effectively, helping lenders reduce financial risks and benefiting all parties involved.

KEYWORDS:

• P2P lending
• feature selection
• decision tree
• modified gray wolf optimization

JEL CLASSIFICATION:

• C61

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Muhammad Sam’an

Muhammad Sam’an received his Master’s degree in Mathematics from the University of Diponegoro in 2016. He works as a lecturer at the Department of Informatics, Universitas Muhammadiyah Semarang, Indonesia. His research interests include fuzzy optimization and machine learning.

Mustafa Mat Deris

Mustafa Mat Deris received the Ph.D. degree from University Putra Malaysia, in 2002. He is currently a Professor of computer science with the Faculty of Computer Science and Information Technology, UTHM, Malaysia and Lecturer in PhD in Information Technology with the Faculty of Business, Management and Information Technology. He published more than 220 papers in journals and conference proceedings with Scopus H-index of 19. He was appointed as a keynote speaker for several conferences and served as a program committee member and a co-organizer for numerous international conferences/workshops. His research interests include distributed databases, rough set theory, soft set theory, and data mining. He was a recipient of ICT Excellent Teacher Award from Malaysian National Computing Confederation (MNCC), in 2006. He has appointed as an Editorial Board Member of Journal of Next Generation Information Technology (JNIT, South Korea), International Journal of Rough Sets and Data Analysis (IGI-Global, USA), and Encyclopedia on Mobile Computing and Commerce (Idea Group, USA); a Guest Editor of International Journal of Biomedical Soft Computing and Human Science for a Special Issue on “Soft Computing Methodologies and Its Applications;” and a Reviewer of several outstanding international journals, such as IEEE Transaction on Parallel and Distributed Computing, Journal of Parallel and Distributed Databases, Journal of Future Generation on Computer Systems, Journal of Information Sciences (Elsevier), and Knowledge-Based Systems (Elsevier).

Farikhin

Farikhin received his Ph.D. from Universiti Malaysia Terengganu, Malaysia, and a Master’s and Bachelor’s degree in Mathematics from Universitas Gadjah Mada, Indonesia. He works as a lecturer at the Department of Mathematics, Universitas Diponegoro, Indonesia. He is interested in numerical optimization, computing, and applied analysis for decision science areas.