Methodological foundations of a universal approach to business communication language analytics based on large language models

Authors

DOI:

https://doi.org/10.26906/EiR.2025.3(98).3919

Keywords:

digital economy, systems analysis, artificial intelligence, natural language processing, language analytics, unit economics, large language models, managerial accounting, machine learning methods

Abstract

The article substantiates the feasibility of using large language models (LLMs) as a fundamentally new tool in business-oriented language analytics tasks. Unlike traditional rule-based approaches and dictionary methods, which are characterized by low flexibility and limited ability for contextual interpretation, as well as classical machine learning algorithms that require large volumes of labeled data and quickly lose relevance in a dynamic business environment, LLMs provide comprehensive multiparametric analysis of communications. Their scientific novelty lies in the ability to integrate semantic, contextual, and emotional dimensions of interaction, while also taking into account industry-specific features without the need to develop narrowly specialized models. The study systematizes applied tasks of language analytics implemented through LLMs, including automatic dialogue summarization, script compliance control, intent identification, thematic tagging, sentiment analysis, lead qualification, upselling and cross-selling opportunity detection, handling of interrupted communications, and customer satisfaction analysis. For each task, universal prompts with dynamic parameters are proposed, enabling adaptation of instructions to the specifics of individual business domains. Particular emphasis is placed on three sectors with critical dependence on language analytics — automotive dealerships, healthcare institutions, and banking services. This approach makes it possible to combine the standardization of analytical procedures with domain-specific detailing for each industry. The practical value of the study lies in the development of a unified methodological framework for integrating LLM-based language analytics into CRM systems, quality control modules, management dashboards, and regulatory reporting. The use of dynamic parameters ensures both universality and scalability of the approach, opening prospects for its application in a wider range of business areas where client communication is a key factor in ensuring unit economics efficiency and enhancing enterprise competitiveness.

Author Biography

  • Dmytro Zhukovskyi, Ukrainian State University of Science and Technologies

    Postgraduate Student

References

Calderon-Monge E. & Ribeiro-Soriano D. The role of digitalization in business and management: a systematic literature review. Review of Managerial Science. 2023. P. 1–43.

Turing A. M. Computing Machinery and Intelligence. Mind. 1950. Vol. 59. № 236. P. 433–460. DOI: https://doi.org/10.1093/mind/LIX.236.433

Church K. and Mercer R. Introduction to the special issue on computational linguistics using large corpora. Computational Linguistics. 19(1). 1993. P. 1-24.

Bahja M. Natural Language Processing Applications in Business. E-Business - Higher Education and Intelligence Applications. 2020. P. 1-25. DOI: https://doi.org/10.5772/intechopen.92203

5.Vaswani A., Shazeer N., Parmar N., Uszkoreit J., Jones L., Gomez A. N., Kaiser Ł. and Polosukhin I. (2017) Attention is All You Need.Advances in Neural Information Processing Systems. № 30. P. 5998-6008.

Kang N., Singh B., Afzal Z. Using rule-based natural language processing to improve disease normalization in biomedical text. Journal of the American Medical Informatics Association. № 20(5). 2013. P. 953–958. DOI: https://doi.org/10.1136/amiajnl-2012-001173

Smith J., Brown P., Taylor R. Rule-based natural language processing for clinical practice. Journal of Clinical Informatics. № 38(4). 2024. P. 332-337.

Finkel J.R., Grenager T., Manning C. Bayesian analysis in natural language processing. Computational Linguistics. № 44(1). 2018. P. 187-218. DOI: https://doi.org/10.1162/COLI_r_00310

Hou Y., Huang J. Natural language processing for social science research: A comprehensive review. Chinese Journal of Sociology. № 11(1). P. 121-157. DOI: https://doi.org/10.1177/2057150X241306780

Ge, Y., Al-Garadi, M.A. and Sarker, A. Data Augmentation with Nearest Neighbor Classifier for Few-Shot Named Entity Recognition. Studies in Health Technology and Informatics. № 314. 2024. P. 220-225. DOI: https://doi.org/10.3233/SHTI231053

Wang J., Li Y., Chen Q. A semantics aware random forest for text classification. Pattern Recognition Letters. № 126. 2019. P. 1061-1070. DOI: https://doi.org/10.1145/3357384.3357891

Goldberg Y. A primer on neural network models for natural language processing. Journal of Artificial Intelligence Research. № 60. 2017 .P. 345-420. DOI: https://doi.org/10.1613/jair.4992

Jayatilleke N., Weerasinghe R., Senanayake N. Advancements in Natural Language Processing for Automatic Text Summarization. International Conference on Computer Systems (ICCS). 2024. P. 74-84. DOI: https://doi.org/10.1109/ICCS62594.2024.10795848

Barr V., Klavans J.L. Verification and Validation of Language Processing Systems. Proceedings of the First Meeting of the North American Chapter of the Association for Computational Linguistics. 2001. P. 6-14.

15. Saranraj R., Vasanthakumar A., Arivoli T. Pretrained Natural Language Processing Model for Intent Recognition. Atlantis Press. P. 66-76.

16. Pan Z., Bai K., Wang Y., Zhou L., Liu X. Improving Open-Domain Dialogue Systems via Multi-Turn Incomplete Utterance Restoration. Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics (NAACL). 2019 P. 1824-1833. DOI: https://doi.org/10.18653/v1/D19-1191

Published

2025-09-30

Issue

Section

Mathematical methods, models and information technologies in economics

How to Cite

Methodological foundations of a universal approach to business communication language analytics based on large language models. (2025). Economics and Region, 3(98), 208-215. https://doi.org/10.26906/EiR.2025.3(98).3919