R.E. Semenov1

1 MIREA – Russian Technological University (RTU MIREA) (Moscow, Russia)
1 9629790@gmail.com

The conceptual graph language model of text understanding is an innovative approach to the analysis and understanding of texts based on the use of conceptual graphs. In this article, the main principles of this model, its advantages and prospects for application in various fields such as machine learning, information retrieval, natural language processing, and others are considered. The aim of the work is to compare the developed model of conceptual graph representation of text information with existing pre-trained models for working with text and identifying the advantages of the developed model.

In the article, the main concepts and principles of the conceptual graph model, as well as examples of its application, will be considered. The advantages and limitations of this model, as well as the prospects for its development and use in various fields will be described. Comparison with already existing pre-trained models is carried out on the basis of neural network training, using the models described in the article. Comparative analysis will allow identifying the percentage ratio of accuracy in determining semantic constructions of different networks and making a conclusion based on the results obtained.

In the article, the conceptual graph language model of text understanding, its main components and operating principles will be described. The process of analyzing text using this model will be considered, and tables with the accuracy of networks of different models on different training sets will be provided.

Special attention is paid to the advantages and limitations of the conceptual graph model, as well as its prospects for development and use in various fields.

In conclusion, the article summarizes the findings and draws conclusions about the significance of the conceptual graph model for text analysis and its potential in identifying the meaning of text when using a small amount of training data. As a result of a series of experiments, it was possible to achieve an accuracy comparable to existing models of pre-trained networks. The accuracy of the conceptual graph network exceeds existing algorithms when trained on the smallest set of 5,000 words. When experimenting with large text sets, the developed model can compete with large language models.

Semenov R.E. Conceptual – graph model of text understanding. Science Intensive Technologies. 2024. V. 25. № 6. P. 34−43. DOI: https://doi.org/10.18127/ j19998465-202406-05 (in Russian)

1. Anferov M.A. Geneticheskij algoritm klasterizacii. Russian Technological Journal. 2019. V. 6(7). P. 134–150 (in Russian).
2. Sorokin A.B., Lobanov D.A. Konceptual'noe proektirovanie intellektual'nyh sistem. Informacionnye tekhnologii. 2018. V. 1(24). P. 3–10 (in Russian).
3. Citul'skij A.M., Ivannikov A.V., Rogov I.S. NLP – obrabotka estestvennyh yazykov. Nauchno-obrazovatel'nyj zhurnal dlya studentov i prepodavatelej «StudNet». 2020. V. 6. P. 467–475 (in Russian).
4. Sorokin A.B., Smol'yaninova V.A. Konceptual'noe proektirovanie ekspertnyh sistem podderzhki prinyatiya reshenij. Informacionnye tekhnologii. 2017. V. 9(23). P. 634–641 (in Russian).
5. Krasnikov K.E. Matematicheskoe modelirovanie nekotoryh social'nyh processov s pomoshch'yu teoretiko-igrovyh podhodov i prinyatie na ih osnove upravlencheskih reshenij. Russian Technological Journal. 2021. V. 9(5). P. 67–83 (in Russian).
6. Zhang, X., Zhao, H., Chen, D.-Y. Semantic Mapping Methods Between Expert View and Ontology View. Journal of Software. 2020. V. 31(9). P. 2855–2882.
7. Citul'skij A.M., Rogov I.S., Ivannikov A.V. Intellektual'nyj analiz teksta. Nauchno-obrazovatel'nyj zhurnal dlya studentov i prepodavatelej «StudNet». 2020. V. 6. P. 476–483 (in Russian).
8. Sorokin A.B., ZHeleznyak L.M., Suprunenko D.V., Holmogorov V.V. Proektirovanie modulej sistemnoj dinamiki v sistemah podderzhki prinyatiya reshenij. Russian Technological Journal. 2022. V. 10(4). P. 18–26 (in Russian).
9. Anferov M.A. Algoritm poiska podkriticheskih putej na setevyh grafikah. Russian Technological Journal. 2023. V. 11(1). P. 60–69 (in Russian).
10. Tomashevskaya V.S., YAkovlev D.A. Sposoby obrabotki nestrukturirovannyh dannyh. Russian Technological Journal. 2021. V. 9(1). P. 7–17 (in Russian).
11. Tatur M.M., Lukashevich M.M., Percev D.Yu., Iskra N.A. Intellektual'nyj analiz dannyh i oblachnye vychisleniya. Doklady BGUIR. 2019. № 6(124). S. 62–71 (in Russian).
12. Minaev V.A., Simonov A.V. Sravnenie modelej-transformerov bert pri vyyavlenii destruktivnogo kontenta v social'nyh media. Informaciya i bezopasnost'. 2022. № 25(3). S. 341–348 (in Russian).
13. Ol'gina I.G., Pronin I.V., Abdrahmanov A.N. Postroenie grafovyh modelej seti citirovaniya nauchnyh publikacij. Sistemy upravleniya, informacionnye tekhnologii i matematicheskoe modelirovanie. 2020. № 1. S. 118–125 (in Russian).
14. Ivanov M.V. Sovremennye nejrosetevye podhody k resheniyu zadach na grafah. Primenenie iskusstvennogo intellekta v informacionno-telekommunikacionnyh sistemah. 2021. S. 90–98 (in Russian).
15. Kochkarov R.A. Issledovanie nekotoryh trudnorazreshimyh zadach na klasse predfraktal'nyh grafov s izmenyaemym traektornym porozhdeniem. Vestnik voronezhskogo gosudarstvennogo universiteta. Seriya: Sistemnyj analiz i informacionnye tekhnologii. 2021. S. 66–82 (in Russian).
16. Bang Liu, Lingfei Wu. Graph Neural Networks in Natural Language Processing. Graph Neural Networks: Foundations, Frontiers, and Applications. 2022. P. 463–481.
17. Luce le Gorrec, Philip A. Knight, Auguste Caen. Learning network embeddings using small graphlets. Social Network Analysis and Mining. 2022. P. 12–20.
18. Haifeng Wang, Jiwei Li, Hua Wu, Eduard Hovy, Yu Sun. Pre-Trained Language Models and Their Applications. Engineering. 2023. V. 25. P. 51–65.
19. Jie Zhou, Ganqu Cui, Shengding Hu, Zhengyan Zhang, Cheng Yang, Zhiyuan Liu, Lifeng Wang, Changcheng Li, Maosong Sun. Graph neural networks: A review of methods and applications. AI Open. 2020. V. 1. P. 57–81.
20. Belov S.D., Zrelova D.P., Zrelov P.V., Koren'kov V.V. Obzor metodov avtomaticheskoj obrabotki tekstov na estestvennom yazyke. Sistemnyj analiz v nauke i obrazovanii. 2020. № 3. S. 8–22 (in Russian).
21. Sadovskaya L.L., Gus'kov A.E., Kosyakov D.V., Muhamediev R.I. Obrabotka tekstov na estestvennom yazyke: obzor publikacij. Iskusstvennyj intellekt i prinyatie reshenij. 2021. № 3. S. 66–86 (in Russian).
22. Gordeeva E.V., Kochkarov R.A., Rylov A.A. Analiz zadachi raspoznavaniya temy teksta s pomoshch'yu mashinnogo obucheniya. Nejrokomp'yutery: razrabotka, primenenie. 2023. T. 25. № 4. S. 7−15 (in Russian).