informИнформатикаInformatics1816-03012617-6963UIIP NASB10.37661/1816-0301-2023-20-3-90-105inform-1261Research ArticleИНТЕЛЛЕКТУАЛЬНЫЕ СИСТЕМЫINTELLIGENT SYSTEMSСемантические модели и средства разработки искусственных нейронных сетей и их интеграции с базами знанийSemantic models and tools for the development of artificial neural networks and their integration into knowledge baseshttps://orcid.org/0000-0002-9606-9541КовалёвМ. В.KovalevM. V.

Ковалёв Михаил Владимирович - исследователь технических наук, кафедра интеллектуальных информационных технологий.

ул. П. Бровки, 6, Минск, 220013

Mikhail V. Kovalev - Researcher of Technical Sciences, the Department of Intelligent Information Technologies, Belarusian State University of Informatics and Radioelectronics.

Brovki P. st., 6, Minsk, 220013

kovalev@bsuir.byБелорусский государственный университет информатики и радиоэлектроникиBelarusian State University of Informatics and Radioelectronics20232909202320390105Copyright © Ковалёв М.В., 20232023Ковалёв М.В.Kovalev M.V.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/1261Цели

Цели. Предлагаются спецификации моделей и средств разработки искусственных нейронных сетей (ИНС) и их интеграции с базами знаний интеллектуальных систем. Актуальность исследования определяется необходимостью решения комплексных задач интеллектуальными системами, алгоритм и методы решения которых отсутствуют в базах знаний интеллектуальных систем.

Методы

Методы. Анализируются четыре уровня интеграции ИНС с базами знаний. В ходе анализа формулируются требования и спецификации к необходимым моделям и средствам разработки и интеграции. Специфицированные на каждом уровне модели и средства включают в себя модели и средства предыдущего уровня. Применение средств рассмотрено на примере решения задачи классификации сущностей базы знаний с помощью графовой нейронной сети.

Результаты

Результаты. Разработаны спецификации модели представления ИНС в базе знаний, агентно-ориентированной модели разработки и интерпретации ИНС, обеспечивающие интеграцию ИНС в базы знаний на всех выделенных уровнях, а также метод классификации сущностей базы знаний с помощью графовой нейронной сети.

Заключение

Заключение. Предложенные модели и средства позволяют интегрировать в базу знаний интеллектуальной системы любые обученные ИНС и использовать их для решения комплексных задач в рамках технологии OSTIS. Становятся возможными проектирование и обучение ИНС на основании как внешних данных, так и фрагментов базы знаний, а также автоматизация процесса разработки ИНС в базе знаний.

Objectives

Objectives. Specifications of models and tools for the development of artificial neural networks (ANNs) and their integration into knowledge bases (KBs) of intelligent systems are being developed. The relevance is determined by the necessity of implementing the possibility to solve complex problems by intelligent systems, which algorithms and methods of solving are not available in the knowledge base of the intelligent system.

Methods

Methods. Four levels of integration of artificial neural networks into knowledge bases are formulated and analyzed. During the analysis the requirements and specifications for required models and tools for the development and integration are formulated. Specified at each level the models and tools include the models and tools of previous level. The application of the tools is considered by the example of solving the problem of classifying the knowledge base entities using a graph neural network.

Results

Results. The specifications of the ANN representation model in the knowledge base, the agent-based model for the development and interpretation of the ANN, which ensures the integration of the ANN into knowledge bases at all selected levels, as well as the method for classifying knowledge base entities using a graph neural network, have been developed.

Conclusion

Conclusion. The developed models and tools allow integrating any trained ANNs into the knowledge base of the intelligent system and using them to solve complex problems within the framework of OSTIS technology. It also becomes possible to design and train ANNs both on the basis of external data and on the basis of fragments of the knowledge base. Automation of ANNs development process in the knowledge base becomes available.

интеграцияметоды решения задачискусственные нейронные сетиграфовые нейронные сетибазы знанийонтологииintegrationproblem-solving methodsartificial neural networksgraph neural networksknowledge basesontologiesReferencesAttention Is All You Need [Electronic resource] / A. Vaswani [et al.]. - 2017. - Mode of access: https://doi.org/10.48550/arXiv.1706.03762. - Date of access: 20.06.2023.Vaswani A., Shazeer N., Parmar N., Uszkoreit J., Jones L., ..., Polosukhin I. Attention Is All You Need, 2017. Available at: https://doi.org/10.48550/arXiv.1706.03762 (accessed 17.06.2023).Training Language Models to Follow Instructions with Human Feedback [Electronic resource] / L. Ouyang [et al.]. - 2022. - Mode of access: https://doi.org/10.48550/arXiv.2203.02155. - Date of access: 20.06.2023.Ouyang L., Wu J., Jiang X., Almeida D., Wainwright C. L., Lowe R. J. Training Language Models to Follow Instructions with Human Feedback, 2022. Available at: https://doi.org/10.48550/arXiv.2203.02155 (accessed 17.06.2023).Radford, A. Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks [Electronic resource] / A. Radford, L. Metz, S. Chintala. - 2016. - Mode of access: https://doi.org/10.48550/arXiv.1511.06434. - Date of access: 20.06.2023.Radford A., Metz L., Chintala S. Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks, 2016. Available at: https://doi.org/10.48550/arXiv.1511.06434 (accessed 17.06.2023).Kejriwal, M. Knowledge graphs: A practical review of the research landscape [Electronic resource] / M. Kejriwal // Applied Sciences. - 2022. - Vol. 13(4), no. 161. - Mode of access: https://doi.org/10.3390/info13040161. - Date of access: 20.06.2023.Kejriwal M. Knowledge graphs: A practical review of the research landscape. Applied Sciences, 2022, vol. 13(4), no. 161. Available at: https://doi.org/10.3390/info13040161 (accessed 17.06.2023).Tsamoura, E. Neural-symbolic integration: a compositional perspective / E. Tsamoura, T. Hospedales, L. Michael // Proceedings of the AAAI Conf. on Artificial Intelligence, held virtually, 2-9 Feb. 2021. - Vol. 35, no. 6. - P. 5051-5060. https://doi.org/10.1609/aaai.v35i6.16639Tsamoura E., Hospedales T., Michael L. Neural-symbolic integration: a compositional perspective. Proceedings of the AAAI Conference on Artificial Intelligence, held virtually, 2-9 February 2021. Vol. 35, no. 6, рр. 5051-5060. https://doi.org/10.1609/aaai.v35i6.16639Deep Learning Meets Knowledge Graphs: A Comprehensive Survey [Electronic resource] / S. Yu [et al.]. - 2022. - Mode of access: https://doi.org/10.21203/rs.3.rs-2021923/v1. - Date of access: 20.06.2023.Yu S., Xu C., Bai X., Bai X., Kuncheerathodi R., ., Xia F. Deep Learning Meets Knowledge Graphs: A Comprehensive Survey, 2022. Available at: https://doi.org/10.21203/rs.3.rs-2021923/v1 (accessed 17.06.2023).Golenkov, V. Next-generation intelligent computer systems and technology of complex support of their life cycle / V. V. Golenkov, N. A. Gulyakina // Открытые семантические технологии проектирования интеллектуальных систем = Open Semantic Technologies for Intelligent Systems (OSTIS-2022) : сб. науч. тр. / редкол.: В. В. Голенков [и др.]. - Минск : БГУИР, 2022. - Вып. 6. - P. 27-40.Golenkov V. V., Gulyakina N. A. Next-generation intelligent computer systems and technology of complex support of their life cycle. Otkrytye semanticheskie tehnologii proektirovanija intellektual'nyh system : sbornik nauchnyh trudov [Open Semantic Technologies for Designing Intelligent Systems (OSTIS-2022) : Research Papers Collection]. Ed. board: V. V. Golenkov, I. S. Azarov, V. A. Golovko, A. N. Gordey, N. A. Guliakina, ., D. V. Shunkevich. Minsk, Belorusskij gosudarstvennyj universitet informatiki i radiojelektroniki, 2022, iss. 6, pp. 27-40.Stokel-Walker, C. AI chatbots are coming to search engines - can you trust the results? [Electronic resource] / C. Stokel-Walker // Nature. - 2023. - Mode of access: https://www.nature.com/articles/d41586-023-00423-4. - Date of access: 20.06.2023.Stokel-Walker C. AI chatbots are coming to search engines - can you trust the results? Nature, 2023. Available at: https://www.nature.com/articles/d41586-023-00423-4 (accessed 17.06.2023).Hoy, M. B. Alexa, Siri, Cortana, and More: An introduction to voice assistants / M. B. Hoy // Medical Reference Services Quarterly. - 2018. - Vol. 37, no. 1. - P. 81-88.Hoy M. B. Alexa, Siri, Cortana, and more: An introduction to voice assistants. Medical Reference Services Quarterly, 2018, vol. 37, no. 1, рр. 81-88.Wiley, V. Computer vision and image processing: A paper review / V. Wiley, T. Lucas // International J. of Artificial Intelligence Research. - 2018 - Vol. 2, no. 1. - P. 28-36. https://doi.org/10.29099/ijair.v2i1.42Wiley V., Lucas T. Computer vision and image processing: A paper review. International Journal of Artificial Intelligence Research, 2018, vol. 2, no. 1, рр. 28-36. https://doi.org/10.29099/ijair.v2i1.42A survey on state-of-the-art techniques for knowledge graphs construction and challenges ahead / A. Hur [et al.] // IEEE Fourth Intern. Conf. on Artificial Intelligence and Knowledge Engineering (AIKE), Laguna Hills, 1-3 Dec. 2021. - Laguna Hills, 2021. - P. 99-103. https://doi.org/10.1109/AIKE52691.2021.00021Hur A., Janjua N., Ahmed M. A survey on state-of-the-art techniques for knowledge graphs construction and challenges ahead. IEEE Fourth International Conference on Artificial Intelligence and Knowledge Engineering (AIKE), Laguna Hills, 1-3 December 2021. Laguna Hills, 2021, рр. 99-103. https://doi.org/10.1109/AIKE52691.2021.00021A neural-symbolic approach to computer vision / A. Kroshchanka [et al.] // Communications in Computer and Information Science : Proceedings of the 11th Intern. Conf. "Open Semantic Technologies for Intelligent Systems" (OSTIS-2021), Minsk, 16-18 Sept. 2021. - Cham : Springer, 2022. - Vol. 1625. - P. 282-309. https://doi.org/10.1007/978-3-031-15882-7_15Kroshchanka A., Golovko V., Mikhno E., Kovalev M., ., Zagorskij A. A neural-symbolic approach to computer vision. Communications in Computer and Information Science : Proceedings of the 11th International Conference "Open Semantic Technologies for Intelligent Systems" (OSTIS-2021), Minsk, 16-18 September 2021. Cham, Springer, 2022, vol. 1625, рр. 282-309. https://doi.org/10.1007/978-3-031-15882-7_15Интеграция искусственных нейронных сетей с базами знаний / В. А. Головко [и др.] // Онтология проектирования. - 2018. - Т. 8, № 3(29). - С. 366-386. https://doi.org/10.18287/2223-9537-2018-8-3-366-386Golovko V., Golenkov V., Ivashenko V., Taberko V., Ivaniuk D., ., Kovalev M. Integration of artificial neural networks with knowledge bases. Ontologija proektirovanija [Design Ontology], 2018, vol. 8, no. 3(29), pp. 366-386. https://doi.org/10.18287/2223-9537-2018-8-3-366-386 (In Russ.).Nielsen, F. A. Inferring visual semantic similarity with deep learning and Wikidata: Introducing / F. A. Nielsen L. K. Hansen // CEUR Workshop Proceedings : Proceedings of the First Workshop on "Deep Learning for Knowledge Graphs and Semantic Technologies" (DL4KGS) co-located with the 15th "Extended Semantic Web Conf." (ESWC 2018), Heraklion, 4 June 2018. - Heraklion, 2018. - Vol. 2106. - P. 56-61.Nielsen F. A., Hansen, L. K. Inferring visual semantic similarity with deep learning and Wikidata: Introducing. CEUR Workshop Proceedings : Proceedings of the First Workshop on "Deep Learning for Knowledge Graphs and Semantic Technologies" (DL4KGS) co-located with the 15th "Extended Semantic Web Conference" (ESWC 2018), Heraklion, 4 June 2018. Heraklion, 2018, vol. 2106, рр. 56-61.Modeling relational data with graph convolutional networks / M. Schlichtkrull [et al.] // Lecture Notes in Computer Science : Proceedings of the 15th Intern. Conf. "European Semantic Web Conf." (ESWC 2018), Heraklion, 3-7 June 2018. - Cham : Springer, 2018. - Vol. 10843. - P. 593-607. https://doi.org/10.1007/978-3-319-93417-4_38Schlichtkrull M., Kipf T. N., Bloem P., van den Berg R., Titov I., Welling M. Modeling relational data with graph convolutional networks. Lecture Notes in Computer Science : Proceedings of the 15th International Conference "European Semantic Web Conference" (ESWC 2018), Heraklion, 3-7 June 2018. Cham, Springer, 2018, vol. 10843, рр. 593-607. https://doi.org/10.1007/978-3-319-93417-4_38Comet, M. Clustering knowledge graphs [Electronic resource] / M. Comet, L. Teresa // Seminar Paper. - 2018. - P. 1-12. - Mode of access: https://doi.org/10.13140/RG.2.2.12494.08008. - Date of access: 20.06.2023.Molinas C., Teresa L. Clustering knowledge graphs. Seminar Paper, 2018, рр. 1-12. Available at: https://doi.org/10.13140/RG.2.2.12494.08008 (accessed 17.06.2023).Saranya, A. A systematic review of explainable artificial intelligence models and applications: Recent developments and future trends / A. Saranya, R. Subhashini // Decision Analytics J. - 2023. - Vol. 7. - P. 100-230. https://doi.org/10.1016/j.dajour.2023.100230Saranya A., Subhashini R. A systematic review of explainable artificial intelligence models and applications: Recent developments and future trends. Decision Analytics Journal, 2023, vol. 7, pp. 100-230. https://doi.org/10.1016/j.dajour.2023.100230Averkin, A. Explainable artificial intelligence: rules extraction from neural networks / A. Averkin, S. Yarushev // Proceedings of the 11th Intern. Conf. on Theory and Application of Soft Computing, Computing with Words and Perceptions and Artificial Intelligence (ICSCCW-2021), Antgalya, 26-27 Aug. 2021. - Antgalya, 2021. - Vol. 362. - P. 102-109. https://doi.org/10.1007/978-3-030-92127-9_17Averkin A., Yarushev S. Explainable artificial intelligence: rules extraction from neural networks. Proceedings of the 11th International Conference on Theory and Application of Soft Computing, Computing with Words and Perceptions and Artificial Intelligence (ICSCCW-2021), Antgalya, 26-27 August 2021. Antgalya, 2021, vol. 362, рр. 102-109. https://doi.org/10.1007/978-3-030-92127-9_17Castelvecchi, D. Can we open the black box of AI? / D. Castelvecchi // Nature. - 2016. - Vol. 538, no. 7623. - P. 20-23.Castelvecchi D. Can we open the black box of AI? Nature, 2016, vol. 538, no. 7623, рр. 20-23.Chomsky, N. The False Promise of ChatGPT / N. Chomsky, I. Roberts, J. Watumull. - Mode of access: https://www.nytimes.com/2023/03/08/opinion/noam-chomsky-chatgpt-ai.html. - Date of access: 20.06.2023.Chomsky N., Roberts I., Watumull J. The False Promise of ChatGPT. Available at: https://www.nytimes.com/2023/03/08/opinion/noam-chomsky-chatgpt-ai.html (accessed 17.06.2023).Технология комплексной поддержки жизненного цикла семантически совместимых интеллектуальных компьютерных систем нового поколения / под общ. ред. В. В. Голенкова. - Минск : Бестпринт, 2023. - Гл. 3.6. - 1064 с.Tehnologija kompleksnoj podderzhki zhiznennogo cikla semanticheski sovmestimyh intellektual'nyh komp'juternyh sistem novogo pokolenija. Technology for Comprehensive Life Cycle Support of Semantically Compatible, Next-generation Intelligent Computer Systems. In V. V. Golenkov (ed.). Minsk, Bestprint, 2023, chap. 3.6, 1064 p. (In Russ.).Kovalev, M. Convergence and integration of artificial neural networks with knowledge bases in next-generation intelligent computer systems / M. V. Kovalev, A. A. Kroshchanka, V. A. Golovko // Открытые семантические технологии проектирования интеллектуальных систем = Open Semantic Technologies for Intelligent Systems (OSTIS-2022) : сб. науч. тр. / редкол.: В. В. Голенков [и др.]. - Минск : БГУИР, 2022. - Вып. 6. - С. 173-186.Kovalev M., Kroshchanka A., Golovko V. Convergence and integration of artificial neural networks with knowledge bases in next-generation intelligent computer systems. Otkrytye semanticheskie tehnologii proektirovanija intellektual'nyh system : sbornik nauchnyh trudov [Open Semantic Technologies for Designing Intelligent Systems (OSTIS-2022) : Research Papers Collection]. Ed. board: V. V. Golenkov, I. S. Azarov, V. A. Golovko, A. N. Gordey, N. A. Guliakina, D. V. Shunkevich. Minsk, Belorusskij gosudarstvennyj universitet informatiki i radiojelektroniki, 2022, iss. 6, pp. 173-186.eXnet: an efficient approach for emotion recognition in the wild / M. N. Riaz [et al.] // Sensors. - 2020. - Vol. 20(4), no. 1087. - P. 1-17. https://doi.org/10.3390/s20041087Riaz M., Shen Y., Sohail M., Guo M. eXnet: an efficient approach for emotion recognition in the wild. Sensors, 2020, Vol. 20(4), no. 1087, рр. 1-17. https://doi.org/10.3390/s20041087Integration of large language models with knowledge bases of intelligent systems / K. Bantsevich [et al.] // Открытые семантические технологии проектирования интеллектуальных систем = Open Semantic Technologies for Intelligent Systems (OSTIS-2022) : сб. науч. тр. / редкол.: В. В. Голенков [и др.]. - Минск : БГУИР, 2023. - Вып. 7. - P. 213-218.Bantsevich K., Kovalev M., Tsishchanka V., Malinovskaya N., Andrushevich A. Integration of large language models with knowledge bases of intelligent systems. Otkrytye semanticheskie tehnologii proektirovanija intellektual'nyh system : sbornik nauchnyh trudov [Open Semantic Technologies for Designing Intelligent Systems (OSTIS-2023) : Research Papers Collection]. Ed. board: V. V. Golenkov, I. S. Azarov, V. A. Golovko, A. N. Gordey, N. A. Guliakina, D. V. Shunkevich. Minsk, Belorusskij gosudarstvennyj universitet informatiki i radiojelektroniki, 2023, iss. 7, pp. 213-218.Xinyi, Z. Capsule graph neural network [Electronic resource] / Z. Xinyi, L. Chen // Intern. Conf. on Learning Representations. - 2019. - Mode of access: https://openreview.net/forum?id=Byl8BnRcYm. - Date of access: 20.06.2023.Xinyi Z., Chen L. Capsule graph neural network. International Conference on Learning Representations, 2019. Available at: https://openreview.net/forum?id=Byl8BnRcYm (accessed 17.06.2023).

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