A Study on the Performance Improvement of Machine LearningClassification Model by Resolving Imbalance in Financial Data: Focusing on Credit Card Accidents
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
The advancement of innovative technologies such as artificial intelligence (AI) and big data has considerably increased in the amount of financial data. However, the quantity of abnormal data belonging to minority classes is relatively low, resulting in an exacerbation of data imbalance phenomenon. Data imbalance significantly impairs the performance of classifiers for minority classes. In this study, classification performance comparison experiments were conducted according to the data imbalance resolution methodology using credit card customer data, one of the financial data. Empirical analysis revealed that as the degree of data imbalance decreased, the performance of the classification model tended to improve, and applying the data imbalance resolution methodology led to performance enhancement. When SMOTETomek was applied, the deep neural network (DNN) showed the highest performance improvement, and the performance of the classification model could be improved using a conditional adversarial generative network (CGAN) for financial data. This study proposed solutions for data imbalance that could be applied to financial data, and the proposed methods contributed to the adoption and development of AI technology in the financial industry by making it more suitable for financial practices.
Abstract
The advancement of innovative technologies such as artificial intelligence (AI) and big data has considerably increased in the amount of financial data. However, the quantity of abnormal data belonging to minority classes is relatively low, resulting in an exacerbation of data imbalance phenomenon. Data imbalance significantly impairs the performance of classifiers for minority classes. In this study, classification performance comparison experiments were conducted according to the data imbalance resolution methodology using credit card customer data, one of the financial data. Empirical analysis revealed that as the degree of data imbalance decreased, the performance of the classification model tended to improve, and applying the data imbalance resolution methodology led to performance enhancement. When SMOTETomek was applied, the deep neural network (DNN) showed the highest performance improvement, and the performance of the classification model could be improved using a conditional adversarial generative network (CGAN) for financial data. This study proposed solutions for data imbalance that could be applied to financial data, and the proposed methods contributed to the adoption and development of AI technology in the financial industry by making it more suitable for financial practices.
참고문헌 (24)
- Das S, "On supervised class-imbalanced learning : An updated perspective and some key challenges", Ieee T Artif Intell, 3, pp.973~993 (2022)
- Parvin H, "A new imbalanced learning and dictions tree method for breast cancer diagnosis", Bionanoscience, 7, pp.673~678 (2013)
- Swana EF, "Tomek link and SMOTE approaches for machine fault classification with an imbalanced dataset", Sensors, 22, pp.3246~ (2022)
- Kumar C, "A hybrid data-driven framework for spam detection in online social network", Procedia Comput Sci, 218, pp.124~132 (2023)
- Chawla NV, "SMOTE : synthetic minority over-sampling technique", J Artif Intell Res, 16, pp.321~357 (2002)
- Buda M, "A systematic study of the class imbalance problem in convolutional neural networks", Neural Networks, 106, pp.249~259 (2018)
- Wilson DL, "Asymptotic properties of nearest neighbor rules using edited data", Ieee T Syst Man Cyb, pp.408~421 (1972)
- Ferri FJ, "Considerations about sample-size sensitivity of a family of edited nearest-neighbor rules", Ieee T Syst Man Cy B, 29, pp.667~672 (1999)
- Tomek I, "Two modifications of cnn", Ieee T Syst Man Cyb, 6, pp.769~772 (1976)
- Elhassan T, "Classification of imbalance data using tomek link(t-link)combined with random under-sampling(rus)as a data reduction method", Global J Technol Optim S, 1, pp.2016~ (2016)
- Kim MJ, "Geometric mean based boosting algorithm with over-sampling to resolve data imbalance problem for bankruptcy prediction", Expert Syst Appl, 42, pp.1074~1082 (2015)
- Goodfellow I, "Generative adversarial networks", Commun ACM, 63, pp.139~144 (2020)
- Mirza M, "Conditional generative adversarial nets", arXiv preprint arXiv:1411.1784, ~ (2014)
- Zheng YJ, "Generative adversarial network based telecom fraud detection at the receiving bank", Neural Networks, 102, pp.78~86 (2018)
- Awan SE, "Imputation of missing data with class imbalance using conditional generative adversarial networks", Neurocomputing, 453, pp.164~171 (2021)
- Hosmer DW, "Applied logistic regression" (2000), Wiley
- Breiman L, "Random forests", Machine learning, 45, pp.5~32 (2001)
- Ke G, "Lightgbm : a highly efficient gradient boosting decision tree", Adv Neural Inf Process Syst, 30 (2017)
- 이정택, "Using Machine Learning Algorithms to Forecast the Optimal Bidding Rate in Apartment Auctions", Quantitative Bio-Science, 40(1), pp.31~37 (2021), 자연과학연구소[DOI]
- Rumelhart DE, "Learning representations by back-propagating errors", Nature, 323, pp.533~536 (1986)
- Wilson RL, "Bankruptcy prediction using neural networks", Decis Support Syst, 11, pp.545~557 (1994)
- Powers DM, "Evaluation: from precision, recall and f-measure to roc, informedness, markedness and correlation", arXiv preprint arXiv:2010.16061, ~ (2020)
- Swets JA, "Measuring the accuracy of diagnostic systems", Science, 240, pp.1285~1293 (1988)
- Greiner M, "Principles and practical application of the receiver-operating characteristic analysis for diagnostic tests", Prev Vet Med, 45, pp.23~41 (2000)