Стекінг нейроподібної структури МПГП з RBF шаром на підставі генерування випадкового кортежу її гіперпараметрів для завдань прогнозування

dc.citation.epage55
dc.citation.issue1
dc.citation.journalTitleУкраїнський журнал інформаційних технологій
dc.citation.spage49
dc.citation.volume3
dc.contributor.affiliationНаціональний університет “Львівська політехніка”
dc.contributor.affiliationLviv Polytechnic National University
dc.contributor.authorТкаченко, Р. О.
dc.contributor.authorІзонін, І. В.
dc.contributor.authorДанилик, В. М.
dc.contributor.authorМихалевич, В. Ю.
dc.contributor.authorTkachenko, R. O.
dc.contributor.authorIzonin, I. V.
dc.contributor.authorDanylyk, V. M.
dc.contributor.authorMykhalevych, V. Yu.
dc.coverage.placenameЛьвів
dc.coverage.placenameLviv
dc.date.accessioned2023-03-23T10:27:06Z
dc.date.available2023-03-23T10:27:06Z
dc.date.created2021-10-10
dc.date.issued2021-10-10
dc.description.abstractПідвищення точності прогнозування засобами штучного інтелекту є важливим завданням у різних галузях промисловості, економіки, медицини. Ансамблеве навчання – один із можливих варіантів досягнення цього. Зокрема, побудова стекінгових моделей на підставі різних методів машинного навчання чи з використанням різних частин наявного набору даних демонструє високу точність прогнозу. Проте потреба правильного підбору членів ансамблю, їх оптимальних параметрів тощо зумовлює необхідність великих часових витрат на підготовку та навчання таких моделей. В роботі запропоновано дещо інший підхід до побудови простого, проте ефективного ансамблевого методу. Розроблено нову модель стекінгу нелінійних нейроподібних структур МПГП, основану на використанні тільки одного типу ШНМ як елементної бази ансамблю та застосуванні однакової для усіх членів ансамблю навчальної вибірки. Такий підхід забезпечує певні переваги порівняно з процедурами побудови ансамблів на підставі різних методів машинного навчання, як мінімум у напрямі підбору оптимальних параметрів для кожного з них. Як основу ансамлювання в нашому випадку використано кортеж випадкових гіперпараметрів для кожного окремого члена ансамблю, тобто навчання кожної комбінованої нейроподібної структури МПГП з додатковим RBF шаром як окремого члена ансамблю здійснюється із використанням різних, випадково вибраних значень центрів RBF та центрів мас. Це забезпечує необхідне різноманіття елементів ансамблю. Експериментальні дослідження щодо ефективності роботи запропонованого ансамблю проведено із використанням реального набору даних. Завдання полягає у прогнозуванні величини медичних страхових виплат на підставі низки незалежних атрибутів. Експериментально визначено оптимальну кількість членів ансамблю, яка забезпечує найвищу точність розв’язання поставленої задачі. Здійснено порівняння результатів роботи запропонованого ансамблю з наявними методами цього класу. Встановлено найвищу точність розробленого ансамблю за задовільної тривалості процедури його навчання.
dc.description.abstractImproving prediction accuracy by artificial intelligence tools is an important task in various industries, economics, medicine. Ensemble learning is one of the possible options to solve this task. In particular, the construction of stacking models based on different machine learning methods, or using different parts of the existing data set demonstrates high prediction accuracy of the. However, the need for proper selection of ensemble members, their optimal parameters, etc., necessitates large time costs for the construction of such models. This paper proposes a slightly different approach to building a simple but effective ensemble method. The authors developed a new model of stacking of nonlinear SGTM neural-like structures, which is based on the use of only one type of ANN as an element base of the ensemble and the use of the same training sample for all members of the ensemble. This approach provides a number of advantages over the procedures for building ensembles based on different machine learning methods, at least in the direction of selecting the optimal parameters for each of them. In our case, a tuple of random hyperparameters for each individual member of the ensemble was used as the basis of ensemble. That is, the training of each combined SGTM neural-like structure with an additional RBF layer, as a separate member of the ensemble occurs using different, randomly selected values of RBF centers and centersfof mass. This provides the necessary variety of ensemble elements. Experimental studies on the effectiveness of the developed ensemble were conducted using a real data set. The task is to predict the amount of health insurance costs based on a number of independent attributes. The optimal number of ensemble members is determined experimentally, which provides the highest prediction accuracy. The results of the work of the developed ensemble are compared with the existing methods of this class. The highest prediction accuracy of the developed ensemble at satisfactory duration of procedure of its training is established.
dc.format.extent49-55
dc.format.pages7
dc.identifier.citationСтекінг нейроподібної структури МПГП з RBF шаром на підставі генерування випадкового кортежу її гіперпараметрів для завдань прогнозування / Р. О. Ткаченко, І. В. Ізонін, В. М. Данилик, В. Ю. Михалевич // Український журнал інформаційних технологій. — Львів : Видавництво Львівської політехніки, 2021. — Том 3. — № 1. — С. 49–55.
dc.identifier.citationenTkachenko R. O., Izonin I. V., Danylyk V. M., Mykhalevych V. Yu. (2021) Stekinh neiropodibnoi struktury MPHP z RBF sharom na pidstavi heneruvannia vypadkovoho kortezhu yii hiperparametriv dlia zavdan prohnozuvannia [Stacking of the SGTM neural-like structure with RBF layer based on generation of a random curtain of its hyperparameters for prediction tasks]. Ukrainian Journal of Information Technology (Lviv), vol. 3, no 1, pp. 49-55 [in Ukrainian].
dc.identifier.doihttps://doi.org/10.23939/ujit2021.03.049
dc.identifier.urihttps://ena.lpnu.ua/handle/ntb/57778
dc.language.isouk
dc.publisherВидавництво Львівської політехніки
dc.publisherLviv Politechnic Publishing House
dc.relation.ispartofУкраїнський журнал інформаційних технологій, 1 (3), 2021
dc.relation.ispartofUkrainian Journal of Information Technology, 1 (3), 2021
dc.relation.references[1] Agarwal, S., & Chowdary, C. R. (2020). A-Stacking and ABagging: Adaptive versions of ensemble learning algorithms for spoof fingerprint detection. Expert Systems with Applications, 146, 113–160. https://doi.org/10.1016/j.eswa.2019.113160
dc.relation.references[2] Boodhun, N., & Jayabalan, M. (2018). Risk prediction in life insurance industry using supervised learning algorithms. Complex & Intelligent Systems, 4(2), 145–154. https://doi.org/10.1007/s40747-018-0072-1
dc.relation.references[3] Chaurasia, V., & Pal, S. (2021). Stacking-Based Ensemble Framework and Feature Selection Technique for the Detection of Breast Cancer. SN Computer Science, 2(2), 67. https://doi.org/10.1007/s42979-021-00465-3
dc.relation.references[4] Feng, D.-C., Liu, Z.-T., Wang, X.-D., Chen, Y., Chang, J.-Q., Wei, D.-F., & Jiang, Z.-M. (2020). Machine learning-based compressive strength prediction for concrete: An adaptive boosting approach. Construction and Building Materials, 230, 117000. https://doi.org/10.1016/j.conbuildmat.2019.117000
dc.relation.references[5] Folberth, C., Elliott, J., Müller, C., Balkovič, J., Chryssanthacopoulos, J., Izaurralde, R. C., Jones, C. D., Khabarov, N., Liu, W., Reddy, A., Schmid, E., Skalský, R., Yang, H., Arneth, A., Ciais, P., Deryng, D., Lawrence, P. J., Olin, S., Pugh, T. A. M., … Wang, X. (2019). Parameterizationinduced uncertainties and impacts of crop management harmonization in a global gridded crop model ensemble. PLOS ONE, 14(9), e0221862. https://doi.org/10.1371/ journal. pone.0221862
dc.relation.references[6] Hassan, A. H. A., & Elfaki, E. (2018). Prediction of Electrical Output Power of Combined Cycle Power Plant Using Regression ANN Model. https://doi.org/10.5281/zenodo.1285164
dc.relation.references[7] Ighalo, J. O., Adeniyi, A. G., & Marques, G. (2020). Application of linear regression algorithm and stochastic gradient descent in a machine – learning environment for predicting biomass higher heating value. Biofuels, Bioproducts and Biorefining, 14(6), 1286–1295. https://doi.org/10.1002/bbb.2140
dc.relation.references[8] Izonin, I., Tkachenko, R., Kryvinska, N., Gregus, M., Tkachenko, P., & Vitynskyi, P. (2019). Committee of SGTM Neural-Like Structures with RBF kernel for Insurance Cost Prediction Task. 2019 IEEE 2nd Ukraine Conference on Electrical and Computer Engineering (UKRCON), 1037–1040. https://doi.org/10.1109/UKRCON.2019.8879905
dc.relation.references[9] Kurz, C. F., Maier, W., & Rink, C. (2020). A greedy stacking algorithm for model ensembling and domain weighting. BMC Research Notes, 13(1), 1–6. https://doi.org/10.1186/s13104-020-4931-7
dc.relation.references[10] Medical Cost Personal Datasets. (n.d.). Retrieved 9 December2018, from https://www.kaggle.com/mirichoi0218/insurance
dc.relation.references[11] Pavlyshenko, B. (2018). Using Stacking Approaches for Machine Learning Models. 2018 IEEE Second International Conference on Data Stream Mining Processing (DSMP), 255–258. https://doi.org/10.1109/DSMP.2018.8478522
dc.relation.references[12] Pham, K., Kim, D., Park, S., & Choi, H. (2021). Ensemble learning-based classification models for slope stability analysis. Catena, 196, 104886. https://doi.org/10.1016/j.catena.2020.104886
dc.relation.references[13] Rocca, J. (2021, March 21). Ensemble methods: Bagging, boosting and stacking. Medium. https://towardsdatascience.com/ensemble-methods-baggingboosting-and-stacking-c9214a10a205
dc.relation.references[14] Salah, M., Altalla, K., Salah, A., & Abu-Naser, S. S. (2018). Predicting Medical Expenses Using Artificial Neural Network. International Journal of Engineering and Information Systems (IJEAIS), 2(10), 7.
dc.relation.references[15] Shaikhina, T., & Khovanova, N. A. (2017). Handling limited datasets with neural networks in medical applications: A small-data approach. Artificial Intelligence in Medicine, 75, 51–63. https://doi.org/10.1016/j.artmed.2016.12.003
dc.relation.references[16] Shakhovska, N., Yakovyna, V., & Kryvinska, N. (2020). An Improved Software Defect Prediction Algorithm Using Selforganizing Maps Combined with Hierarchical Clustering and Data Preprocessing. In S. Hartmann, J. Küng, G. Kotsis, A.M. Tjoa, & I. Khalil (Eds.), Database and Expert Systems Applications, 414–424. Springer International Publishing. https://doi.org/10.1007/978-3-030-59003-1_27
dc.relation.references[17] Teslyuk, V., Kazarian, A., Kryvinska, N., & Tsmots, I. (2021). Optimal Artificial Neural Network Type Selection Method for Usage in Smart House Systems. Sensors, 21(1), 47. https://doi.org/10.3390/s21010047
dc.relation.references[18] Tkachenko, R., & Izonin, I. (2019). Model and Principles for the Implementation of Neural-Like Structures Based on Geometric Data Transformations. In Z. Hu, S. Petoukhov, I. Dychka, & M. He (Eds.). Advances in Computer Science for Engineering and Education. Vol. 754, 578–587. Springer International Publishing. https://doi.org/10.1007/978-3-319-91008-6_58
dc.relation.references[19] Tkachenko, R., Izonin, I., Vitynskyi, P., Lotoshynska, N., & Pavlyuk, O. (2018). Development of the Non-Iterative Supervised Learning Predictor Based on the Ito Decomposition and SGTM Neural-Like Structure for Managing Medical Insurance Costs. Data, 3(4), 46. https://doi.org/10.3390/data3040046
dc.relation.references[20] Tkachenko, R., Kutucu, H., Izonin, I., Doroshenko, A., & Tsymbal, Y. (n.d.). Non-Iterative Neural-Like Predictor for Solar Energy in Libya. 11.
dc.relation.references[21] Tkachenko, R., Tkachenko, P., Izonin, I., Vitynskyi, P., Kryvinska, N., & Tsymbal, Y. (2019). Committee of the Combined RBF-SGTM Neural-Like Structures for Prediction Tasks. In I. Awan, M. Younas, P. Ünal, & M. Aleksy (Eds.). Mobile Web and Intelligent Information Systems, 267–277. Springer International Publishing. https://doi.org/10.1007/978-3-030-27192-3_21
dc.relation.references[22] Tsmots, I., & Skorokhoda, O. (2010). Methods and VLSIstructures for neural element implementation. 2010 Proceedings of VIth International Conference on Perspective Technologies and Methods in MEMS Design, 135–135.
dc.relation.references[23] Tsmots, I., Skorokhoda, O., & Rabyk, V. (2016). Structure and Software Model of a Parallel-Vertical Multi-Input Adder for FPGA Implementation, 158–160. https://doi.org/10.1109/STC-CSIT.2016.7589894
dc.relation.references[24] Tsmots, I., Teslyuk, V., & Vavruk, I. (2013). Hardware and software tools for motion control of mobile robotic system. 2013 12th International Conference on the Experience of Designing and Application of CAD Systems in Microelectronics (CADSM), 368–368.
dc.relation.references[25] Xiao, Y., Wu, J., Lin, Z., & Zhao, X. (2018). A deep learning-based multi-model ensemble method for cancer prediction. Computer Methods and Programs in Biomedicine, 153, 1–9. https://doi.org/10.1016/j.cmpb.2017.09.005
dc.relation.referencesen[1] Agarwal, S., & Chowdary, C. R. (2020). A-Stacking and ABagging: Adaptive versions of ensemble learning algorithms for spoof fingerprint detection. Expert Systems with Applications, 146, 113–160. https://doi.org/10.1016/j.eswa.2019.113160
dc.relation.referencesen[2] Boodhun, N., & Jayabalan, M. (2018). Risk prediction in life insurance industry using supervised learning algorithms. Complex & Intelligent Systems, 4(2), 145–154. https://doi.org/10.1007/s40747-018-0072-1
dc.relation.referencesen[3] Chaurasia, V., & Pal, S. (2021). Stacking-Based Ensemble Framework and Feature Selection Technique for the Detection of Breast Cancer. SN Computer Science, 2(2), 67. https://doi.org/10.1007/s42979-021-00465-3
dc.relation.referencesen[4] Feng, D.-C., Liu, Z.-T., Wang, X.-D., Chen, Y., Chang, J.-Q., Wei, D.-F., & Jiang, Z.-M. (2020). Machine learning-based compressive strength prediction for concrete: An adaptive boosting approach. Construction and Building Materials, 230, 117000. https://doi.org/10.1016/j.conbuildmat.2019.117000
dc.relation.referencesen[5] Folberth, C., Elliott, J., Müller, C., Balkovič, J., Chryssanthacopoulos, J., Izaurralde, R. C., Jones, C. D., Khabarov, N., Liu, W., Reddy, A., Schmid, E., Skalský, R., Yang, H., Arneth, A., Ciais, P., Deryng, D., Lawrence, P. J., Olin, S., Pugh, T. A. M., … Wang, X. (2019). Parameterizationinduced uncertainties and impacts of crop management harmonization in a global gridded crop model ensemble. PLOS ONE, 14(9), e0221862. https://doi.org/10.1371/ journal. pone.0221862
dc.relation.referencesen[6] Hassan, A. H. A., & Elfaki, E. (2018). Prediction of Electrical Output Power of Combined Cycle Power Plant Using Regression ANN Model. https://doi.org/10.5281/zenodo.1285164
dc.relation.referencesen[7] Ighalo, J. O., Adeniyi, A. G., & Marques, G. (2020). Application of linear regression algorithm and stochastic gradient descent in a machine – learning environment for predicting biomass higher heating value. Biofuels, Bioproducts and Biorefining, 14(6), 1286–1295. https://doi.org/10.1002/bbb.2140
dc.relation.referencesen[8] Izonin, I., Tkachenko, R., Kryvinska, N., Gregus, M., Tkachenko, P., & Vitynskyi, P. (2019). Committee of SGTM Neural-Like Structures with RBF kernel for Insurance Cost Prediction Task. 2019 IEEE 2nd Ukraine Conference on Electrical and Computer Engineering (UKRCON), 1037–1040. https://doi.org/10.1109/UKRCON.2019.8879905
dc.relation.referencesen[9] Kurz, C. F., Maier, W., & Rink, C. (2020). A greedy stacking algorithm for model ensembling and domain weighting. BMC Research Notes, 13(1), 1–6. https://doi.org/10.1186/s13104-020-4931-7
dc.relation.referencesen[10] Medical Cost Personal Datasets. (n.d.). Retrieved 9 December2018, from https://www.kaggle.com/mirichoi0218/insurance
dc.relation.referencesen[11] Pavlyshenko, B. (2018). Using Stacking Approaches for Machine Learning Models. 2018 IEEE Second International Conference on Data Stream Mining Processing (DSMP), 255–258. https://doi.org/10.1109/DSMP.2018.8478522
dc.relation.referencesen[12] Pham, K., Kim, D., Park, S., & Choi, H. (2021). Ensemble learning-based classification models for slope stability analysis. Catena, 196, 104886. https://doi.org/10.1016/j.catena.2020.104886
dc.relation.referencesen[13] Rocca, J. (2021, March 21). Ensemble methods: Bagging, boosting and stacking. Medium. https://towardsdatascience.com/ensemble-methods-baggingboosting-and-stacking-P.9214a10a205
dc.relation.referencesen[14] Salah, M., Altalla, K., Salah, A., & Abu-Naser, S. S. (2018). Predicting Medical Expenses Using Artificial Neural Network. International Journal of Engineering and Information Systems (IJEAIS), 2(10), 7.
dc.relation.referencesen[15] Shaikhina, T., & Khovanova, N. A. (2017). Handling limited datasets with neural networks in medical applications: A small-data approach. Artificial Intelligence in Medicine, 75, 51–63. https://doi.org/10.1016/j.artmed.2016.12.003
dc.relation.referencesen[16] Shakhovska, N., Yakovyna, V., & Kryvinska, N. (2020). An Improved Software Defect Prediction Algorithm Using Selforganizing Maps Combined with Hierarchical Clustering and Data Preprocessing. In S. Hartmann, J. Küng, G. Kotsis, A.M. Tjoa, & I. Khalil (Eds.), Database and Expert Systems Applications, 414–424. Springer International Publishing. https://doi.org/10.1007/978-3-030-59003-1_27
dc.relation.referencesen[17] Teslyuk, V., Kazarian, A., Kryvinska, N., & Tsmots, I. (2021). Optimal Artificial Neural Network Type Selection Method for Usage in Smart House Systems. Sensors, 21(1), 47. https://doi.org/10.3390/s21010047
dc.relation.referencesen[18] Tkachenko, R., & Izonin, I. (2019). Model and Principles for the Implementation of Neural-Like Structures Based on Geometric Data Transformations. In Z. Hu, S. Petoukhov, I. Dychka, & M. He (Eds.). Advances in Computer Science for Engineering and Education. Vol. 754, 578–587. Springer International Publishing. https://doi.org/10.1007/978-3-319-91008-6_58
dc.relation.referencesen[19] Tkachenko, R., Izonin, I., Vitynskyi, P., Lotoshynska, N., & Pavlyuk, O. (2018). Development of the Non-Iterative Supervised Learning Predictor Based on the Ito Decomposition and SGTM Neural-Like Structure for Managing Medical Insurance Costs. Data, 3(4), 46. https://doi.org/10.3390/data3040046
dc.relation.referencesen[20] Tkachenko, R., Kutucu, H., Izonin, I., Doroshenko, A., & Tsymbal, Y. (n.d.). Non-Iterative Neural-Like Predictor for Solar Energy in Libya. 11.
dc.relation.referencesen[21] Tkachenko, R., Tkachenko, P., Izonin, I., Vitynskyi, P., Kryvinska, N., & Tsymbal, Y. (2019). Committee of the Combined RBF-SGTM Neural-Like Structures for Prediction Tasks. In I. Awan, M. Younas, P. Ünal, & M. Aleksy (Eds.). Mobile Web and Intelligent Information Systems, 267–277. Springer International Publishing. https://doi.org/10.1007/978-3-030-27192-3_21
dc.relation.referencesen[22] Tsmots, I., & Skorokhoda, O. (2010). Methods and VLSIstructures for neural element implementation. 2010 Proceedings of VIth International Conference on Perspective Technologies and Methods in MEMS Design, 135–135.
dc.relation.referencesen[23] Tsmots, I., Skorokhoda, O., & Rabyk, V. (2016). Structure and Software Model of a Parallel-Vertical Multi-Input Adder for FPGA Implementation, 158–160. https://doi.org/10.1109/STC-CSIT.2016.7589894
dc.relation.referencesen[24] Tsmots, I., Teslyuk, V., & Vavruk, I. (2013). Hardware and software tools for motion control of mobile robotic system. 2013 12th International Conference on the Experience of Designing and Application of CAD Systems in Microelectronics (CADSM), 368–368.
dc.relation.referencesen[25] Xiao, Y., Wu, J., Lin, Z., & Zhao, X. (2018). A deep learning-based multi-model ensemble method for cancer prediction. Computer Methods and Programs in Biomedicine, 153, 1–9. https://doi.org/10.1016/j.cmpb.2017.09.005
dc.relation.urihttps://doi.org/10.1016/j.eswa.2019.113160
dc.relation.urihttps://doi.org/10.1007/s40747-018-0072-1
dc.relation.urihttps://doi.org/10.1007/s42979-021-00465-3
dc.relation.urihttps://doi.org/10.1016/j.conbuildmat.2019.117000
dc.relation.urihttps://doi.org/10.1371/
dc.relation.urihttps://doi.org/10.5281/zenodo.1285164
dc.relation.urihttps://doi.org/10.1002/bbb.2140
dc.relation.urihttps://doi.org/10.1109/UKRCON.2019.8879905
dc.relation.urihttps://doi.org/10.1186/s13104-020-4931-7
dc.relation.urihttps://www.kaggle.com/mirichoi0218/insurance
dc.relation.urihttps://doi.org/10.1109/DSMP.2018.8478522
dc.relation.urihttps://doi.org/10.1016/j.catena.2020.104886
dc.relation.urihttps://towardsdatascience.com/ensemble-methods-baggingboosting-and-stacking-c9214a10a205
dc.relation.urihttps://doi.org/10.1016/j.artmed.2016.12.003
dc.relation.urihttps://doi.org/10.1007/978-3-030-59003-1_27
dc.relation.urihttps://doi.org/10.3390/s21010047
dc.relation.urihttps://doi.org/10.1007/978-3-319-91008-6_58
dc.relation.urihttps://doi.org/10.3390/data3040046
dc.relation.urihttps://doi.org/10.1007/978-3-030-27192-3_21
dc.relation.urihttps://doi.org/10.1109/STC-CSIT.2016.7589894
dc.relation.urihttps://doi.org/10.1016/j.cmpb.2017.09.005
dc.rights.holder© Національний університет „Львівська політехніка“, 2021
dc.subjectстекінг
dc.subjectансамблеве навчання
dc.subjectнейроподібні структури
dc.subjectрадіально-базисні функції
dc.subjectмедичні страхові виплати
dc.subjectstacking
dc.subjectensemble learning
dc.subjectneural-like structures
dc.subjectradial-basic functions
dc.subjectmedical insurance costs
dc.subject.udc004.89
dc.titleСтекінг нейроподібної структури МПГП з RBF шаром на підставі генерування випадкового кортежу її гіперпараметрів для завдань прогнозування
dc.title.alternativeStacking of the SGTM neural-like structure with RBF layer based on generation of a random curtain of its hyperparameters for prediction tasks
dc.typeArticle

Files

Original bundle

Now showing 1 - 1 of 1
Thumbnail Image
Name:
2021v3n1_Tkachenko_R_O-Stacking_of_the_SGTM_49-55.pdf
Size:
1.17 MB
Format:
Adobe Portable Document Format

License bundle

Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.87 KB
Format:
Plain Text
Description: