informИнформатикаInformatics1816-03012617-6963UIIP NASB10.37661/1816-0301-2023-20-1-55-74inform-1228Research ArticleИНТЕЛЛЕКТУАЛЬНЫЕ СИСТЕМЫINTELLIGENT SYSTEMSКлассификация займов c использованием логистической регрессииLoan classification using logistic regressionhttps://orcid.org/0000-0003-1142-3992БегунковВ. И.BehunkouU. I.

Бегунков Владимир Иванович, магистр технических наук

ул. Сурганова, 6, Минск, 220012

Uladzimir I. Behunkou, M. Sc. (Eng.)

 

vbegunkov@gmail.comhttps://orcid.org/0000-0003-0832-0829КовалевМ. Я.KovalyovM. Y.

Ковалев Михаил Яковлевич, член-корреспондент НАН Беларуси, доктор физико-математических наук, профессор

ул. Сурганова, 6, Минск, 220012

Mikhail Y. Kovalyov, Corresponding Member of the National Academy of Sciences of Belarus, D. Sc. (Eng.)

 

kovalyov_my@newman.bas-net.byОбъединенный институт проблем информатики Национальной академии наук БеларусиThe United Institute of Informatics Problems of the National Academy of Sciences of Belarus2023290320232015574Copyright © Бегунков В.И., Ковалев М.Я., 20232023Бегунков В.И., Ковалев М.Я.Behunkou U.I., Kovalyov M.Y.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://inf.grid.by/jour/article/view/1228Цели

Цели. Решение задачи классификации займов имеет большое значение для финансовых институтов, которые должны эффективно распределять денежные активы между субъектами в рамках предоставления финансовых услуг. Поэтому финансовым организациям необходим инструмент наиболее точного определения надежных заемщиков. Одним из инструментов принятия таких решений служит машинное обучение. Целью работы является анализ возможности эффективного применения логистической регрессии для решения задачи классификации займов.

Метод

Метод. На основе алгоритма логистической регрессии с использованием исторических данных по выданным займам рассчитываются следующие метрики: стоимостная функция, Accuracy, Precision, Recall и мера . Полиномиальная регрессия и метод главных компонент применяются для определения оптимального набора входных данных для исследуемого алгоритма логистической регрессии.

Результаты

Результаты. Оценено влияние нормализации данных на конечный результат, дана оценка влияния сбалансированности целевых значений, рассчитано оптимальное граничное значение для алгоритма логистической регрессии, рассмотрено влияние увеличения входных показателей посредством заполнения отсутствующих значений и использования полиномов разной степени. Имеющийся набор входных показателей проанализирован на избыточность.

Заключение

Заключение. Результаты исследований подтверждают, что применение алгоритма логистической регрессии для решения задач классификации займов является целесообразным. Данный алгоритм позволяет быстро получить работающий инструмент классификации займов. 

Objectives

Objectives. The studied problem of loan classification is particularly important for financial institutions, which must efficiently allocate monetary assets between entities as part of the provision of financial services. Therefore, it is more important than ever for financial institutions to be able to identify reliable borrowers as accurately as possible. At the same time, machine learning is one of the tools for making such decisions. The aim of this work is to analyze the possibility of efficient use of logistic regression for solving the task of loan  classification.

Methods

Methods. Based on the logistic regression algorithm using historical data on loans issued, the following  metrics are calculated: cost function, Accuracy, Precision, Recall и  score. Polynomial regression and  principal component analysis are used to determine the optimal set of input data for the being studied logistic regression algorithm.

Results

Results. The impact of data normalization on the final result is estimated, the optimal regularization parameter for solving this problem is determined, the impact of the balance of target values is assessed, the optimal  boundary value for the logistic regression algorithm is calculated, the influence of increasing input indicators by means of filling in missing values and using polynomials of different degrees is considered and the existing set of input indicators is analyzed for redundancy.

Conclusion

Conclusion. The research results confirm that the application of the logistic regression algorithm for solving loan classification problems is appropriate. The use of this algorithm allows to get quickly a working loan  classification tool. 

классификация займаскоринглогистическая регрессиямашинное обучениенормализация данныхloan classificationscoringlogistic regressionmachine learningdata normalizationReferencesGerhard F., Harlalka A., Suvanam R. The coming opportunity in consumer lending. McKinsey Quarterly, 2021. Available at: https://www.mckinsey.com/business-functions/risk-and-resilience/our-insights/the-comingopportunity-in-consumer-lending (accessed 01.05.2021).Gerhard F., Harlalka A., Suvanam R. The coming opportunity in consumer lending. McKinsey Quarterly, 2021. Available at: https://www.mckinsey.com/business-functions/risk-and-resilience/our-insights/the-comingopportunity-in-consumer-lending (accessed 01.05.2021).Hand D. J., Henley W. E. Statistical classification methods in consumer credit scoring: a review. Journal of the Royal Statistical Society: Series A (Statistics in Society), 1997, vol. 160, no. 3, pp. 523–541.Hand D. J., Henley W. E. Statistical classification methods in consumer credit scoring: a review. Journal of the Royal Statistical Society: Series A (Statistics in Society), 1997, vol. 160, no. 3, pp. 523–541.Baesens B., Van Gestel T., Viaene S., Stepanova S., Suykens J., Vanthienen J. Benchmarking state-of-theart classification algorithms for credit scoring. Journal of the Operational Research Society, 2003, vol. 54, no. 6, pp. 627–635.Baesens B., Van Gestel T., Viaene S., Stepanova S., Suykens J., Vanthienen J. Benchmarking state-of-theart classification algorithms for credit scoring. Journal of the Operational Research Society, 2003, vol. 54, no. 6, pp. 627–635.Lessmann S., Baesens B., Seow H.-V., Thomas L. C. Benchmarking state-of-the-art classification algorithms for credit scoring: an update of research. European Journal of Operational Research, 2015, vol. 247, no. 1, pp. 124–136.Lessmann S., Baesens B., Seow H.-V., Thomas L. C. Benchmarking state-of-the-art classification algorithms for credit scoring: an update of research. European Journal of Operational Research, 2015, vol. 247, no. 1, pp. 124–136.Shalev-Shwartz S., Ben-David S. Understanding Machine Learning: From Theory to Algorithms. Cambridge University Press, 2014, pp. 125, 126–127.Shalev-Shwartz S., Ben-David S. Understanding Machine Learning: From Theory to Algorithms. Cambridge University Press, 2014, pp. 125, 126–127.Geron A. Hands-on Machine Learning with Scikit-Learn, Keras, and TensorFlow, 2nd edition. O’Reilly Media, 2019, pp. 144–149.Geron A. Hands-on Machine Learning with Scikit-Learn, Keras, and TensorFlow, 2nd edition. O’Reilly Media, 2019, pp. 144–149.Murphy K. P. Machine Learning: A Probabilistic Perspective (Adaptive Computation and Machine Learning Series). The MIT Press, 2012, pp. 225–227, 387–407.Murphy K. P. Machine Learning: A Probabilistic Perspective (Adaptive Computation and Machine Learning Series). The MIT Press, 2012, pp. 225–227, 387–407.Harrington P. Machine Learning in Action, 1st edition. Manning Publication Co, 2012, pp. 86–91, 148, 269–279.Harrington P. Machine Learning in Action, 1st edition. Manning Publication Co, 2012, pp. 86–91, 148, 269–279.Fawcett T. An introduction to ROC analysis. Pattern Recognition Letters, 2006, vol. 27, no. 8, pp. 861–874.Fawcett T. An introduction to ROC analysis. Pattern Recognition Letters, 2006, vol. 27, no. 8, pp. 861–874.Metz C. E. Basic principles of ROC analysis. Seminars in Nuclear Medicine, 1978, vol. 8, no. 4, pp. 283–298.Metz C. E. Basic principles of ROC analysis. Seminars in Nuclear Medicine, 1978, vol. 8, no. 4, pp. 283–298.Kelleher J. D., Namee B. M., D’Arcy A. Fundamentals of Machine Learning for Predictive Data Analytics, 1st edition. The MIT Press, 2015, pp. 142–143, 539.Kelleher J. D., Namee B. M., D’Arcy A. Fundamentals of Machine Learning for Predictive Data Analytics, 1st edition. The MIT Press, 2015, pp. 142–143, 539.Gerhard F., Harlalka A., Suvanam R. The coming opportunity in consumer lending. McKinsey Quarterly, 2021. Available at: https://www.mckinsey.com/business-functions/risk-and-resilience/our-insights/the-comingopportunity-in-consumer-lending (accessed 01.05.2021).Gerhard F., Harlalka A., Suvanam R. The coming opportunity in consumer lending. McKinsey Quarterly, 2021. Available at: https://www.mckinsey.com/business-functions/risk-and-resilience/our-insights/the-comingopportunity-in-consumer-lending (accessed 01.05.2021).Hand D. J., Henley W. E. Statistical classification methods in consumer credit scoring: a review. Journal of the Royal Statistical Society: Series A (Statistics in Society), 1997, vol. 160, no. 3, pp. 523–541.Hand D. J., Henley W. E. Statistical classification methods in consumer credit scoring: a review. Journal of the Royal Statistical Society: Series A (Statistics in Society), 1997, vol. 160, no. 3, pp. 523–541.Baesens B., Van Gestel T., Viaene S., Stepanova S., Suykens J., Vanthienen J. Benchmarking state-of-theart classification algorithms for credit scoring. Journal of the Operational Research Society, 2003, vol. 54, no. 6, pp. 627–635.Baesens B., Van Gestel T., Viaene S., Stepanova S., Suykens J., Vanthienen J. Benchmarking state-of-theart classification algorithms for credit scoring. Journal of the Operational Research Society, 2003, vol. 54, no. 6, pp. 627–635.Lessmann S., Baesens B., Seow H.-V., Thomas L. C. Benchmarking state-of-the-art classification algorithms for credit scoring: an update of research. European Journal of Operational Research, 2015, vol. 247, no. 1, pp. 124–136.Lessmann S., Baesens B., Seow H.-V., Thomas L. C. Benchmarking state-of-the-art classification algorithms for credit scoring: an update of research. European Journal of Operational Research, 2015, vol. 247, no. 1, pp. 124–136.Shalev-Shwartz S., Ben-David S. Understanding Machine Learning: From Theory to Algorithms. Cambridge University Press, 2014, pp. 125, 126–127.Shalev-Shwartz S., Ben-David S. Understanding Machine Learning: From Theory to Algorithms. Cambridge University Press, 2014, pp. 125, 126–127.Geron A. Hands-on Machine Learning with Scikit-Learn, Keras, and TensorFlow, 2nd edition. O’Reilly Media, 2019, pp. 144–149.Geron A. Hands-on Machine Learning with Scikit-Learn, Keras, and TensorFlow, 2nd edition. O’Reilly Media, 2019, pp. 144–149.Murphy K. P. Machine Learning: A Probabilistic Perspective (Adaptive Computation and Machine Learning Series). The MIT Press, 2012, pp. 225–227, 387–407.Murphy K. P. Machine Learning: A Probabilistic Perspective (Adaptive Computation and Machine Learning Series). The MIT Press, 2012, pp. 225–227, 387–407.Harrington P. Machine Learning in Action, 1st edition. Manning Publication Co, 2012, pp. 86–91, 148, 269–279.Harrington P. Machine Learning in Action, 1st edition. Manning Publication Co, 2012, pp. 86–91, 148, 269–279.Fawcett T. An introduction to ROC analysis. Pattern Recognition Letters, 2006, vol. 27, no. 8, pp. 861–874.Fawcett T. An introduction to ROC analysis. Pattern Recognition Letters, 2006, vol. 27, no. 8, pp. 861–874.Metz C. E. Basic principles of ROC analysis. Seminars in Nuclear Medicine, 1978, vol. 8, no. 4, pp. 283–298.Metz C. E. Basic principles of ROC analysis. Seminars in Nuclear Medicine, 1978, vol. 8, no. 4, pp. 283–298.Kelleher J. D., Namee B. M., D’Arcy A. Fundamentals of Machine Learning for Predictive Data Analytics, 1st edition. The MIT Press, 2015, pp. 142–143, 539.Kelleher J. D., Namee B. M., D’Arcy A. Fundamentals of Machine Learning for Predictive Data Analytics, 1st edition. The MIT Press, 2015, pp. 142–143, 539.

The authors declare that there are no conflicts of interest present.