Алгоритм построения динамически стабильных деревьев классификации в условиях нестационарности финансово-экономических параметров объектов
Ключевые слова:
решающие деревья, классификация, динамическая стабильность алгоритмов машинного обучения, регуляризация
Аннотация
Описан оригинальный алгоритм построения решающих деревьев для задачи классификации, учитывающий стабильности целевой переменной и нестационарный характер данных в листьях. Дан обзор существующих решений, в рамках которого выявлены недостатки известных подходов. Даны основные принципы работы алгоритма, его ограничения и преимущества, особенности работы с данными.
Литература
1. Moslehi S. e. a. Application of Machine Learning Models Based on Decision Trees in Classifying the Factors Affecting Mortality of COVID-19 Patients in Hamadan, Iran // BMC Medical Informatics and Decision Making. 2022. V. 22(1). Pp. 1—12.
2. Zhang Z. Applications of the Decision Tree in Business Field // Proc. III Intern. Conf. Economic Management and Cultural Industry. 2021
3. Alvares C.A. e. a. Decision tree Application to Predict and Spatialize the Wood Productivity Probabilities of Eucalyptus Plantations // Forests. 2023. V. 14(7). P. 1334.
4. Charbuty B., Abdulazeez A.M. Classification Based on Decision Tree Algorithm for Machine Learning // J. Appl. Sci. and Technol. Trends. 2021. V. 2(01). Pp. 20—28.
5. Jankowski D., Jackowski K., Cyganek B. Learning Decision Trees from Data Streams with Concept Drift // Proc. Computer Sci. 2016. V. 80. Pp. 1682—1691.
6. Темкин И.О., Клебанов Д.А., Дерябин С.А., Конов И.С. Построение интеллектуальной геоинформационной системы горного предприятия с использованием методов прогнозной аналитики // Горный информационно-аналитический бюллетень. 2020. № 3. С. 114—125.
7. Корольков Т.А. и др. Построение модели идентификации деятельности работников строительного комплекса // Инженерно-строительный вестник Прикаспия. 2022. № 2(40). С. 130—135.
8. Rzazade U. e. a. A. On the Issue of the Creation and Functioning of Energy Efficiency Management Systems for Technological Processes of Mining Enterprises // Energies. 2023. V. 16(13). P. 4878.
9. Bertsimas D., Digalakis V. Improving Stability in Decision Tree Models [Электрон ресурс] https://doi.org/10.48550/arxiv.2305.17299 (дата обращения 20.05.2024).
10. Portoleau T., Artigues C., Guillaume R. Robust Decision Trees for the Multi-mode Project Scheduling Problem with a Resource Investment Objective and Uncertain Activity Duration // European J. Operational Research. 2024. V. 312(2). Pp. 525—540.
11. Wilton J., Ye N. Robust Loss Functions for Training Decision Trees with Noisy Labels [Электрон ресурс] (дата обращения 20.05.2024).
12. Azad M., Chikalov I., Hussain S., Moshkov M. Entropy-based Greedy Algorithm for Decision Trees Using Hypotheses // Entropy. 2021. V. 23(7). P. 808.
13. Farris F.A. The Gini Index and Measures of Inequality // The American Mathematical Monthly. 2010. V. 117(10). P. 851.
14. Yurdakul B., Naranjo J.D. Statistical Properties of the Population Stability Index // J. Risk Model Validation. 2020. V. 14(4). Pp. 89—100.
15. Przybyła–Kasperek M., Aning S. Stop Criterion in Building Decision Trees with Bagging Method for Dispersed Data // Proc. Computer Sci. 2021. V. 192. Pp. 3560—3569.
16. Eftekhari M., Mehrpooya A., Saberi-Movahed F., Torra V. Comparing Different Stopping Criteria for Fuzzy Decision Tree Induction Through IDFID3 // Studies in Fuzziness and Soft Computing. 2022. V. 416. Pp. 81—92.
17. Rokach L., Maimon O. Decision Trees // Springer eBooks. N.Y.: Springer, 2006. Pp. 165—192.
18. Cui S., Sudjianto A., Zhang A., Li R. Enhancing Robustness of Gradient-boosted Decision Trees through One-hot Encoding and Regularization [Электрон ресурс] https://doi.org/10.48550/arxiv.2304.13761 (дата обращения 20.05.2024).
19. Weed D.L. Weight of Evidence: a Review of Concept and Methods // Risk Analysis. 2005. V. 25(6). Pp. 1545—1557.
20. Bramer M. Avoiding Overfitting of Decision Trees // Undergraduate Topics in Computer Sci. N.-Y.: Springer, 2016. Pp. 121—136.
---
Для цитирования: Мышлянов А.В., Темкин И.О. Алгоритм построения динамически стабильных деревьев классификации в условиях нестационарности финансово-экономических параметров объектов // Вестник МЭИ. 2025. № 1. С. 147—156. DOI: 10.24160/1993-6982-2025-1-147-156
---
Конфликт интересов: авторы заявляют об отсутствии конфликта интересов
#
1. Moslehi S. e. a. Application of Machine Learning Models Based on Decision Trees in Classifying the Factors Affecting Mortality of COVID-19 Patients in Hamadan, Iran. BMC Medical Informatics and Decision Making. 2022;22(1):1—12.
2. Zhang Z. Applications of the Decision Tree in Business Field. Proc. III Intern. Conf. Economic Management and Cultural Industry. 2021
3. Alvares C.A. e. a. Decision tree Application to Predict and Spatialize the Wood Productivity Probabilities of Eucalyptus Plantations. Forests. 2023;14(7):1334.
4. Charbuty B., Abdulazeez A.M. Classification Based on Decision Tree Algorithm for Machine Learning. J. Appl. Sci. and Technol. Trends. 2021;2(01):20—28.
5. Jankowski D., Jackowski K., Cyganek B. Learning Decision Trees from Data Streams with Concept Drift. Proc. Computer Sci. 2016;80:1682—1691.
6. Temkin I.O., Klebanov D.A., Deryabin S.A., Konov I.S. Postroenie Intellektual'noy Geoinformatsionnoy Sistemy Gornogo Predpriyatiya s Ispol'zovaniem Metodov Prognoznoy Analitiki. Gornyy Informatsionno-analiticheskiy Byulleten'. 2020;3:114—125. (in Russian).
7. Korol'kov T.A. i dr. Postroenie Modeli Identifikatsii Deyatel'nosti Rabotnikov Stroitel'nogo Kompleksa. Inzhenerno-stroitel'nyy Vestnik Prikaspiya. 2022;2(40):130—135. (in Russian).
8. Rzazade U. e. a. A. On the Issue of the Creation and Functioning of Energy Efficiency Management Systems for Technological Processes of Mining Enterprises. Energies. 2023;16(13):4878.
9. Bertsimas D., Digalakis V. Improving Stability in Decision Tree Models [Elektron Resurs] https://doi.org/10.48550/arxiv.2305.17299 (Data Obrashcheniya 20.05.2024).
10. Portoleau T., Artigues C., Guillaume R. Robust Decision Trees for the Multi-mode Project Scheduling Problem with a Resource Investment Objective and Uncertain Activity Duration. European J. Operational Research. 2024;312(2):525—540.
11. Wilton J., Ye N. Robust Loss Functions for Training Decision Trees with Noisy Labels [Elektron Resurs] (Data Obrashcheniya 20.05.2024).
12. Azad M., Chikalov I., Hussain S., Moshkov M. Entropy-based Greedy Algorithm for Decision Trees Using Hypotheses. Entropy. 2021;23(7):808.
13. Farris F.A. The Gini Index and Measures of Inequality. The American Mathematical Monthly. 2010;117(10):851.
14. Yurdakul B., Naranjo J.D. Statistical Properties of the Population Stability Index. J. Risk Model Validation. 2020;14(4):89—100.
15. Przybyła–Kasperek M., Aning S. Stop Criterion in Building Decision Trees with Bagging Method for Dispersed Data. Proc. Computer Sci. 2021;192:3560—3569.
16. Eftekhari M., Mehrpooya A., Saberi-Movahed F., Torra V. Comparing Different Stopping Criteria for Fuzzy Decision Tree Induction Through IDFID3. Studies in Fuzziness and Soft Computing. 2022;416:81—92.
17. Rokach L., Maimon O. Decision Trees. Springer eBooks. N.Y.: Springer, 2006:165—192.
18. Cui S., Sudjianto A., Zhang A., Li R. Enhancing Robustness of Gradient-boosted Decision Trees through One-hot Encoding and Regularization [Elektron Resurs] https://doi.org/10.48550/arxiv.2304.13761 (Data Obrashcheniya 20.05.2024).
19. Weed D.L. Weight of Evidence: a Review of Concept and Methods. Risk Analysis. 2005;25(6):1545—1557.
20. Bramer M. Avoiding Overfitting of Decision Trees. Undergraduate Topics in Computer Sci. N.-Y.: Springer, 2016:121—136
---
For citation: Myshlyanov A.V., Temkin I.O. An Algorithm for Constructing Dynamically Stable Classification Trees under the Conditions of Unsteady Financial and Economic Parameters of Objects. Bulletin of MPEI. 2025;1:147—156. (in Russian). DOI: 10.24160/1993-6982-2025-1-147-156
---
Conflict of interests: the authors declare no conflict of interest
2. Zhang Z. Applications of the Decision Tree in Business Field // Proc. III Intern. Conf. Economic Management and Cultural Industry. 2021
3. Alvares C.A. e. a. Decision tree Application to Predict and Spatialize the Wood Productivity Probabilities of Eucalyptus Plantations // Forests. 2023. V. 14(7). P. 1334.
4. Charbuty B., Abdulazeez A.M. Classification Based on Decision Tree Algorithm for Machine Learning // J. Appl. Sci. and Technol. Trends. 2021. V. 2(01). Pp. 20—28.
5. Jankowski D., Jackowski K., Cyganek B. Learning Decision Trees from Data Streams with Concept Drift // Proc. Computer Sci. 2016. V. 80. Pp. 1682—1691.
6. Темкин И.О., Клебанов Д.А., Дерябин С.А., Конов И.С. Построение интеллектуальной геоинформационной системы горного предприятия с использованием методов прогнозной аналитики // Горный информационно-аналитический бюллетень. 2020. № 3. С. 114—125.
7. Корольков Т.А. и др. Построение модели идентификации деятельности работников строительного комплекса // Инженерно-строительный вестник Прикаспия. 2022. № 2(40). С. 130—135.
8. Rzazade U. e. a. A. On the Issue of the Creation and Functioning of Energy Efficiency Management Systems for Technological Processes of Mining Enterprises // Energies. 2023. V. 16(13). P. 4878.
9. Bertsimas D., Digalakis V. Improving Stability in Decision Tree Models [Электрон ресурс] https://doi.org/10.48550/arxiv.2305.17299 (дата обращения 20.05.2024).
10. Portoleau T., Artigues C., Guillaume R. Robust Decision Trees for the Multi-mode Project Scheduling Problem with a Resource Investment Objective and Uncertain Activity Duration // European J. Operational Research. 2024. V. 312(2). Pp. 525—540.
11. Wilton J., Ye N. Robust Loss Functions for Training Decision Trees with Noisy Labels [Электрон ресурс] (дата обращения 20.05.2024).
12. Azad M., Chikalov I., Hussain S., Moshkov M. Entropy-based Greedy Algorithm for Decision Trees Using Hypotheses // Entropy. 2021. V. 23(7). P. 808.
13. Farris F.A. The Gini Index and Measures of Inequality // The American Mathematical Monthly. 2010. V. 117(10). P. 851.
14. Yurdakul B., Naranjo J.D. Statistical Properties of the Population Stability Index // J. Risk Model Validation. 2020. V. 14(4). Pp. 89—100.
15. Przybyła–Kasperek M., Aning S. Stop Criterion in Building Decision Trees with Bagging Method for Dispersed Data // Proc. Computer Sci. 2021. V. 192. Pp. 3560—3569.
16. Eftekhari M., Mehrpooya A., Saberi-Movahed F., Torra V. Comparing Different Stopping Criteria for Fuzzy Decision Tree Induction Through IDFID3 // Studies in Fuzziness and Soft Computing. 2022. V. 416. Pp. 81—92.
17. Rokach L., Maimon O. Decision Trees // Springer eBooks. N.Y.: Springer, 2006. Pp. 165—192.
18. Cui S., Sudjianto A., Zhang A., Li R. Enhancing Robustness of Gradient-boosted Decision Trees through One-hot Encoding and Regularization [Электрон ресурс] https://doi.org/10.48550/arxiv.2304.13761 (дата обращения 20.05.2024).
19. Weed D.L. Weight of Evidence: a Review of Concept and Methods // Risk Analysis. 2005. V. 25(6). Pp. 1545—1557.
20. Bramer M. Avoiding Overfitting of Decision Trees // Undergraduate Topics in Computer Sci. N.-Y.: Springer, 2016. Pp. 121—136.
---
Для цитирования: Мышлянов А.В., Темкин И.О. Алгоритм построения динамически стабильных деревьев классификации в условиях нестационарности финансово-экономических параметров объектов // Вестник МЭИ. 2025. № 1. С. 147—156. DOI: 10.24160/1993-6982-2025-1-147-156
---
Конфликт интересов: авторы заявляют об отсутствии конфликта интересов
#
1. Moslehi S. e. a. Application of Machine Learning Models Based on Decision Trees in Classifying the Factors Affecting Mortality of COVID-19 Patients in Hamadan, Iran. BMC Medical Informatics and Decision Making. 2022;22(1):1—12.
2. Zhang Z. Applications of the Decision Tree in Business Field. Proc. III Intern. Conf. Economic Management and Cultural Industry. 2021
3. Alvares C.A. e. a. Decision tree Application to Predict and Spatialize the Wood Productivity Probabilities of Eucalyptus Plantations. Forests. 2023;14(7):1334.
4. Charbuty B., Abdulazeez A.M. Classification Based on Decision Tree Algorithm for Machine Learning. J. Appl. Sci. and Technol. Trends. 2021;2(01):20—28.
5. Jankowski D., Jackowski K., Cyganek B. Learning Decision Trees from Data Streams with Concept Drift. Proc. Computer Sci. 2016;80:1682—1691.
6. Temkin I.O., Klebanov D.A., Deryabin S.A., Konov I.S. Postroenie Intellektual'noy Geoinformatsionnoy Sistemy Gornogo Predpriyatiya s Ispol'zovaniem Metodov Prognoznoy Analitiki. Gornyy Informatsionno-analiticheskiy Byulleten'. 2020;3:114—125. (in Russian).
7. Korol'kov T.A. i dr. Postroenie Modeli Identifikatsii Deyatel'nosti Rabotnikov Stroitel'nogo Kompleksa. Inzhenerno-stroitel'nyy Vestnik Prikaspiya. 2022;2(40):130—135. (in Russian).
8. Rzazade U. e. a. A. On the Issue of the Creation and Functioning of Energy Efficiency Management Systems for Technological Processes of Mining Enterprises. Energies. 2023;16(13):4878.
9. Bertsimas D., Digalakis V. Improving Stability in Decision Tree Models [Elektron Resurs] https://doi.org/10.48550/arxiv.2305.17299 (Data Obrashcheniya 20.05.2024).
10. Portoleau T., Artigues C., Guillaume R. Robust Decision Trees for the Multi-mode Project Scheduling Problem with a Resource Investment Objective and Uncertain Activity Duration. European J. Operational Research. 2024;312(2):525—540.
11. Wilton J., Ye N. Robust Loss Functions for Training Decision Trees with Noisy Labels [Elektron Resurs] (Data Obrashcheniya 20.05.2024).
12. Azad M., Chikalov I., Hussain S., Moshkov M. Entropy-based Greedy Algorithm for Decision Trees Using Hypotheses. Entropy. 2021;23(7):808.
13. Farris F.A. The Gini Index and Measures of Inequality. The American Mathematical Monthly. 2010;117(10):851.
14. Yurdakul B., Naranjo J.D. Statistical Properties of the Population Stability Index. J. Risk Model Validation. 2020;14(4):89—100.
15. Przybyła–Kasperek M., Aning S. Stop Criterion in Building Decision Trees with Bagging Method for Dispersed Data. Proc. Computer Sci. 2021;192:3560—3569.
16. Eftekhari M., Mehrpooya A., Saberi-Movahed F., Torra V. Comparing Different Stopping Criteria for Fuzzy Decision Tree Induction Through IDFID3. Studies in Fuzziness and Soft Computing. 2022;416:81—92.
17. Rokach L., Maimon O. Decision Trees. Springer eBooks. N.Y.: Springer, 2006:165—192.
18. Cui S., Sudjianto A., Zhang A., Li R. Enhancing Robustness of Gradient-boosted Decision Trees through One-hot Encoding and Regularization [Elektron Resurs] https://doi.org/10.48550/arxiv.2304.13761 (Data Obrashcheniya 20.05.2024).
19. Weed D.L. Weight of Evidence: a Review of Concept and Methods. Risk Analysis. 2005;25(6):1545—1557.
20. Bramer M. Avoiding Overfitting of Decision Trees. Undergraduate Topics in Computer Sci. N.-Y.: Springer, 2016:121—136
---
For citation: Myshlyanov A.V., Temkin I.O. An Algorithm for Constructing Dynamically Stable Classification Trees under the Conditions of Unsteady Financial and Economic Parameters of Objects. Bulletin of MPEI. 2025;1:147—156. (in Russian). DOI: 10.24160/1993-6982-2025-1-147-156
---
Conflict of interests: the authors declare no conflict of interest
Опубликован
2024-10-24
Выпуск
Раздел
Системный анализ, управление и обработка информации (2.3.1)
