In this competitive world, companies need to manage their arrears of debtors to survive. Big companies monthly face thousands of debt cases from their customers and if these debtor customers do not pay their debts within the specified period, these cases will enter the legal stage as legal debt collection that is handled by experienced lawyers. At the legal stage, the cases are sent to the contracted legal debt collection agencies to start a lawsuit. Therefore, assigning which case to which legal debt collection agency is a significant and critical issue in the company's success in getting its debts in the legal stage. So, this study aims to find the legal debt collection agency, which has more capability to close the assigned debt case by predicting case closing probability with machine learning techniques based on past historical data. To predict case closing probability, 8 machine learning algorithms, Catboost Classifier, Extreme Gradient Boosting Classifier, Gradient Boosting Classifier etc., are applied to the processed dataset. The results show us Catboost Classifier has the best accuracy performance 0.87 accuracy. Also, the results show us boosting type ensemble learning algorithms have better performance than other algorithms. Finally, we tune hyper-parameters of Catboost classifier to get better accuracy in the modeling and applied k-fold cross-validation for testing the model's testing stability.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.