E.V. Romanova1, O.D. Mironova2, V.A. Malekova3

1,3 Financial University under the Government of the Russian Federation (Moscow, Russia)
2 Moscow Institute of Physics and Technology (National Research University) (Dolgoprudny, Moscow Region)
1 EkVRomanova@fa.ru, 3 VAMalekova@fa.ru

The user support portal of the Business Incubator of the Financial University over government of Russian Federation is a vital tool for facilitating interaction between users and the organization’s service staff. Users seek information on the operations of the Business Incubator, assistance with business ideas, document preparation, and consultations regarding taxation, accounting, and legal aspects of entrepreneurship.

Each day, the user support service receives numerous requests that require prompt processing by staff. Managing this flow of inquiries is challenging due to limited processing time, the necessity of delivering timely responses, and the classification of requests for redirection to appropriate specialists.

To effectively address the distribution of user inquiries, modern automated message processing methods can be advantageous. Machine learning, part of artificial intelligence, enables the creation of solutions based on natural language processing algorithms, which are particularly suited for directing user requests efficiently.

Integrating machine learning algorithms into business processes can be optimally achieved using user-friendly tools. One such solution is a Telegram chatbot equipped with an embedded machine learning algorithm. Developing and implementing this chatbot will streamline communication processes within the organization, reduce response times, and alleviate the workload on operators.

The proposed solution offers several benefits, including round-the-clock support service availability, a reduction in operator errors, faster task completion, lower operational costs, better optimization of human resources, and an improved user experience.

By creating solutions grounded in natural language processing and integrating them with user interfaces, the interaction processes between the company and its clients can be optimized, significantly enhancing customer experience. The primary goal of this initiative is to develop and implement a Telegram chatbot featuring embedded machine learning capabilities to enhance user interactions with the Business Incubator's support portal. This innovative approach aims to improve efficiency, provide timely assistance, and ultimately lead to a more satisfactory and productive engagement for users.

Romanova E.V., Mironova O.D., Malekova V.A. Developing a chatbot for a user support portal using machine learning methods. Nonlinear World. 2025. V. 23. № 3. P. 54–65. DOI: https:// doi.org/10.18127/ j20700970-202503-07 (In Russian)

1. Chizhova I.A., Kublik E.I., Chipchagov M.S., Labincev A.I. Povyshenie effektivnosti fil'tracii spama na osnove metodov mashinnogo obucheniya v soobshcheniyah razlichnoj prirody. Nejrokomp'yutery: razrabotka, primenenie: mezhdunarodnyj nauchno-tekhnicheskij zhurnal. 2022. T. 24. № 5. S. 5–18. DOI 10.18127/j19998554-202205-01 (In Russian).
2. Koroteev M.V. Obzor sovremennyh metodov obrabotki estestvennyh yazykov na osnove mashinnogo obucheniya. Hi-Hume Jornal – zhurnal vysokih gumanitarnyh tekhnologij. 2024. № 4. (In Russian)
3. Prokopchina S.V., Ryabov P.E., Shchetinin E.Yu. Mashinnoe obuchenie svertochnoj nejronnoj seti na osnove regulyariziruyushchego bajesovskogo podhoda. Nejrokomp'yutery: razrabotka, primenenie. 2024. T. 26. № 3. S. 36–44. DOI: 10.18127/j19998 554-202403-04 (In Russian).
4. Koroteev M.V. Obzor nekotoryh sovremennyh tendencij v tekhnologii mashinnogo obucheniya. E-Management. 2018. T. 1. № 1. S. 26–35. DOI: 10.26425/2658-3445-2018-1-26-35 (In Russian).
5. Kalazhokov Z.H. Osobennosti postroeniya nejronnyh setej, vliyayushchie na ih kachestvo i sposobnost' k obucheniyu. Avtomatizaciya. Sovremennye tekhnologii. 2023. T. 77. № 12. S. 539–543 (In Russian).
6. Maksyutin P.A., Shul'zhenko S.N. Obzor metodov klassifikacii tekstov s pomoshch'yu mashinnogo obucheniya. Inzhenernyj vestnik dona. 2022. № 12 (96). S. 1–9 (In Russian).
7. Al-Hawari F., Barham H. A machine learning based help desk system for IT service management. Journal of King Saud University – Computer and Information Sciences. 2021. V. 33. № 6. P. 702–718. DOI:10.1016/j.jksuci.2019.04.001.
8. Angari A., Marcuzzo M., Schiavinato M., Gasparetto A., Albarelli A. Ticket automation: An insight into current research with applications to multi-level classification scenarios. Expert Systems with Applications. 2023. V. 225. URL: https://www.sciencedirect.com/science/ article/pii/S0957417423 004864?via%3Dihub. Data obrashcheniya: 17.09.2024, doi:10.1016/j.eswa.2023.119984.
9. Korolev S.A., Polupanov A.A. Dialogovaya sistema sluzhby podderzhki na osnove glubokogo obucheniya. Prikladnaya matematika: sovremennye problemy matematiki, informatiki i modelirovaniya: Sb. trudov Materialy II Vseros. nauch.-prakt. konf. (Krasnodar, 22–25 aprelya 2020 g.) Krasnodar, 2020. S. 341–345 (In Russian).
10. Glazkova A.V. Sravnenie nejrosetevyh modelej dlya klassifikacii tekstovyh fragmentov, soderzhashchih biograficheskuyu informaciyu. Mezhdunarodnyj zhurnal Programmnye produkty i sistemy. 2019. № 2. S. 263–267. DOI:10.15827/0236-235X.126. 263-267 (In Russian).
11. Lutfullaeva M., Medvedeva M., Komotskiy E., Spasov K. Optimization of Sentiment Analysis Methods for classifying text comments of bank customers. IFAC-PapersOnLine. 2018. V. 51. 32. P. 55–60. URL: https://www.sciencedirect.com/science/article/pii/ S2405896318330507. Data obrashcheniya: 17.09.2024, doi:10.1016/j.ifacol. 2018.11.353.
12. Ngai E., Li Xiu, Chau D. Application of data mining techniques in customer relationship management: A literature review and classification 1 March 2009 Business, Computer Science DOI:10.1016/j.eswa.2008.02.021.
13. Kochkarov A.A., Makrushin S.V., Kamenchuk V.E., Blohin N. V. Eksperimental'naya teoriya grafov i algoritmy analiza setevyh modelej: uchebnoe posobie. M.: KnoRus. 2024. 160 s. (In Russian).
14. Malekova V.A., Romanova E.V. Generaciya voprosov na estestvennom yazyke s pomoshch'yu nejrosetej. Problemy ekonomiki i yuridicheskoj praktiki. 2022. T. 18. № 2. S. 235–239 (In Russian).