Machine learning in R&D of linguistic projects with artificial intelligence in crisis times - Abstract - Keywords - Introduction

Aug 4, 2026

Machine learning (ML) serves as a core component of artificial intelligence (AI) and a key method for linguistic research during crises. Classical approaches support interpretability and efficiency for structured data, while neural methods handle large-scale and unstructured information more effectively. In unstable conditions, ML supports automation, adaptability, and rapid text analysis for tasks such as NLP, machine translation, and corpus-based forecasting of language trends. Despite issues like data quality, bias, and limited resources, hybrid classical-neural integration improves both accuracy and resilience, sustaining continuity of linguistic R&D.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages5

Reading Time13 min

AccessFree to download

Introduction

Machine learning paradigms in AI

The Main Part

Directions of ML in linguistic research

Benefits of ML in crisis contexts

Використання функціоналу Cisco Cloudlock з організаційними заходами дозволяєстворити ефективну систему безпеки менеджменту безпеки хмарного середовища . Отже менеджмент інформаційної безпеки в хмарних середовищах є одниміз найважливіших напрямів корпоративного управління безпекою . Використанняхмарного сервісу Cisco Cloudlock для менеджменту інформаційної безпекизабезпечує компаніям ефективний захист від загроз, контроль та управлінняінцидентами, контроль доступу до ресурсів та даних і відповідність стандартам івимогам . Таким чином використання даного сервісу в системах менеджментуінформаційно безпеки дозволяє компаніям знизити ризики, підвищити довіруклієнтів та забезпечити стабільність бізнес процесів у хмарних середовищах . Список використаних джерел 1. Назаренко Д.М . Огляд проблем захисту даних у хмарних технологіяхСучасні інформаційні системи та технології : матеріали 3-ї науково-практичноїконференції . 4 – 16 Листопада 2019, Харків, Україна С . 170–172. 2. Cisco Cloudlock Documentation [Електронний ресурс]–URL:[https://docs.umbrella.com/](https://docs.umbrella.com/)[ ](https://docs.umbrella.com/)cloudlock-documentation/docs/building-different-types-of-event-policies (дата звернення: 10.08.2025). 3. Технологія забезпечення кібербезпеки хмарного середовища на базі рішенняСisco Сloudlock [Електронний ресурс]–URL: [https://journals.dut.edu.ua/index.php/](https://journals.dut.edu.ua/index.php/)[ ](https://journals.dut.edu.ua/index.php/)[dataprotect/article/view/2663/2557](dataprotect/article/view/2663/2557) (дата звернення: 10.08.2025) . UDC 004.94:81’32 Machine learning in R&D of linguistic projects with artificial intelligence in crisis times Svitlana Krasnyuk Kyiv National University of Technologies and Design, Kyiv [https://orcid.org/0000-0002-5987-8681](https://orcid.org/0000-0002-5987-8681) Abstract. Machine learning (ML) is a core element of artificial intelligence (AI) and a key tool for linguistic research during crises. Classical ML ensures interpretability and efficiency for structured data, while neural network approaches are effective for large-scale and unstructured information. In unstable contexts, ML enables automation, adaptability, and rapid analysis of texts. Its applications include natural language processing, machine translation, and corpus-based forecasting of linguistic trends. Despite challenges such as data quality, bias, and resource limitations, hybrid integration of classical and neural methods provides both accuracy and resilience. Thus, ML strengthens the stability and continuity of linguistic R&D projects in times of crisis. Keywords: computational linguistics, machine learning, artificial intelligence. Introduction. Machine learning (ML) is the core of modern artificial intelligence (AI) technologies [1]. In general, ML can be divided into two main paradigms: classical (statistical, symbolic) machine learning [2], [3] and neural network machine learning (based on shallow [4], [5] and deep [6] neural network architectures). Classical [7] and neural network [8] machine learning are two fundamental paradigms, each with its own principles, methods, data and computational resource requirements, application areas, and strengths and weaknesses [9]. Classical ML is suitable for structured and small data with transparent algorithms, while neural network ML excels in analyzing & analytics of BIG, multitarget, high-dimensional and unstructured data [10]. Modern solutions in AI are often built on a combination of both paradigms to ensure efficiency, accuracy, and adaptability. Modern linguistic research faces rapidly changing conditions of the external environment: economic instability, socio-political crises, technological transformations. In such conditions, traditional methods of text and language data analysis often become insufficiently flexible and require large human resources. It is machine learning that provides opportunities for automation, adaptation and accel

FAQs

How do classical and neural approaches differ in machine learning for linguistic research?

Classical ML emphasizes interpretability and efficiency for structured, small datasets, while neural network ML is better for analyzing large, high-dimensional, unstructured data and analytics.

Which linguistic research tasks are highlighted as main applications of machine learning?

The document highlights natural language processing, machine translation/automatic translation, and analysis of large text corpora, including forecasting and trend detection.

What advantages does machine learning provide in crisis and unstable environments?

ML enables automation, adaptability, and faster analysis, improves accuracy via hidden pattern detection and prediction, and reduces human error while supporting scalability.

Aug 4, 2026

Upload by Clementine