CatBoost for Predicting Imbalanced Data with Distribution Confusion

Zonglin He

doi:10.47297/taposatWSP2633-456922.20220305

PDF
导出
分享
收藏
专辑

CatBoost for Predicting Imbalanced Data with Distribution Confusion

3卷第5期

浏览量: 0 Downloads: 9

CatBoost for Predicting Imbalanced Data with Distribution Confusion
CatBoost for Predicting Imbalanced Data with Distribution Confusion
School of Artificial Intelligence, Guilin University of Electronic Technology, Guilin Guangxi 541004, P.R. China
2022年3卷第5期页码：112-116
纸质出版日期： 2022-05-25 ，
DOI：10.47297/taposatWSP2633-456922.20220305
Accepted：

Scan QR Code

(2022). CatBoost for Predicting Imbalanced Data with Distribution Confusion. 科技理论与实践（英文版）, 3(5), 112-116.

He Zonglin.CatBoost for Predicting Imbalanced Data with Distribution Confusion[J].Theory and Practice of Science and Technology,2022,03(05):112-116.
(2022). CatBoost for Predicting Imbalanced Data with Distribution Confusion. 科技理论与实践（英文版）, 3(5), 112-116. DOI： 10.47297/taposatWSP2633-456922.20220305.

He Zonglin.CatBoost for Predicting Imbalanced Data with Distribution Confusion[J].Theory and Practice of Science and Technology,2022,03(05):112-116. DOI： 10.47297/taposatWSP2633-456922.20220305.

摘要

Abstract

As the economy improves

people have become more and more conscious of overspending

and this has led to the development of the loan industry. Banks and platforms such as Jingdong Gold Bar and Ant Credit Pay are providing loans while relying on interest to make a profit. At the same time

loans solve the dilemma of customers' lack of funds. The loan business achieves a win-win situation for both the lending platform and the customers. However

some customers violate the agreement and do not repay their loans

which can cause huge financial losses to the lending platform. Therefore

it is of great significance to identify potential defaulting users using default prediction models. This paper cleans the data with SmoteENN

which not only balance the data but also solves the problem that the performance of the classification model degrades due to irregular sample distribution. After that

Logistic Regression

GBDT

XGBoost

LightGBM

and CatBoost are used to make predictions respectively. By analyzing the classification model evaluation metrics such as Recall

F1-score

and AUC

it was found that CatBoost with SmoteENN showed the best results on the dataset.

关键词

Keywords

SmoteENNCatBoost

references

Fitria, N., Sari, R. L. (2012) .Analisis Kebijakan Pemberian Kredit Dan Pengaruh Non Performing Loan Terhadap Loan To Deposit Ratio Pada Pt. Bank Rakyat Indonesia (Persero), Tbk Cabang Rantau, Aceh Tamiang.(Periode 2007-2011). Ekonomi dan Keuangan, 1: 88-101.

Emekter, R., Tu, Y., Jirasakuldech, B., Lu, M. (2015) .Evaluating credit risk and loan performance in online Peer-to-Peer (P2P) lending. Applied Economics, 47: 54-70.

Zhang, C., Liu, C., Zhang, X., Almpanidis, G. (2017). An up-to-date comparison of state-of-the-art classification algorithms. Expert Systems with Applications, 82: 128-150.

Chen, T., Guestrin, C. (2016).Xgboost: A scalable tree boosting system. In: Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining. San Francisco. pp. 785-794.

Ma, X., Sha, J., Wang, D., Yu, Y., Yang, Q., Niu, X. (2018).Study on a prediction of P2P network loan default based on the machine learning LightGBM and XGboost algorithms according to different high dimensional data cleaning. Electronic Commerce Research and Applications, 31: 24-39.

Prokhorenkova, L., Gusev, G., Vorobev, A., Dorogush, A. V., Gulin, A. (2018).Catboost: Unbiased boosting with categorical features. In: Advances in Neural Information Processing Systems. Montreal. pp. 6638-6648.

Chawla, N. V., Bowyer, K. W., Hall, L. O., Kegelmeyer, W. P. (2002). SMOTE: synthetic minority over-sampling technique. Journal of artificial intelligence research, 16: 321-357.

Wilson, D. L. (1972) .Asymptotic properties of nearest neighbor rules using edited data. IEEE Transactions on Systems, Man, and Cybernetics, SMC-2: 408-421.

Batista, G. E., Prati, R. C., Monard, M. C. (2004).A study of the behavior of several methods for balancing machine learning training data. ACM SIGKDD explorations newsletter, 6: 20-29.

文章被引用时，请邮件提醒。

Submit

No data

Related Author

No data

Related Institution

No data

⁰