ANALYSIS OF THE APPLICABILITY OF MODELING FOR MONITORING EDUCATIONAL ORGANIZATION’S REPUTATIONAL RISKS IN THE DIGITAL ENVIRONMENT
Section: GENERAL SECTION
Proceedings: 7TH BETANCOURT INTERNATIONAL ENGINEERING FORUM
Authors:
1.
Saint-Petersburg University of State Fire Service of EMERCOM of Russia
(Faculty of Advanced Training, Graduate Student)
Russian Federation
Russian Federation
Type:
Сonference article
Pages:
from 259
to 264
Published:
25.06.2026
Subject area:
UDC 004.056
Language:
Russian
Keywords:
topic modeling, online reputation, social media analysis, trust, artificial intelligence
Abstract:
The article examines the systematic monitoring of public opinion on education in the context of the digital transformation of society and the increasing mediatization of the information environment. It emphasizes the strategic importance of analyzing stakeholders’ perceptions of educational systems and organizations to strengthen trust and improve the effectiveness of managerial decisions. The study reviews modern text-processing methods, including Aspect-Based Sentiment Analysis (ABSA), topic modeling (LDA, NMF, BERTopic, HDP, CTM), and key-aspect extraction, which allow for the identification of latent discourse themes, the evaluation of the sentiment associated with specific aspects of educational activities, and the detection of potential reputational risks. A comparative analysis of topic modeling techniques is provided, highlighting their features when working with different types of texts and streaming data. The practical significance of the article lies in justifying the use of natural language processing tools for systematic monitoring of the online reputation of educational organizations, early detection of informational threats, and the formation of an evidence base for managerial decisions, thereby enhancing the resilience of educational institutions in the media environment.
The article examines the systematic monitoring of public opinion on education in the context of the digital transformation of society and the increasing mediatization of the information environment. It emphasizes the strategic importance of analyzing stakeholders’ perceptions of educational systems and organizations to strengthen trust and improve the effectiveness of managerial decisions. The study reviews modern text-processing methods, including Aspect-Based Sentiment Analysis (ABSA), topic modeling (LDA, NMF, BERTopic, HDP, CTM), and key-aspect extraction, which allow for the identification of latent discourse themes, the evaluation of the sentiment associated with specific aspects of educational activities, and the detection of potential reputational risks. A comparative analysis of topic modeling techniques is provided, highlighting their features when working with different types of texts and streaming data. The practical significance of the article lies in justifying the use of natural language processing tools for systematic monitoring of the online reputation of educational organizations, early detection of informational threats, and the formation of an evidence base for managerial decisions, thereby enhancing the resilience of educational institutions in the media environment.
Keywords:
topic modeling, online reputation, social media analysis, trust, artificial intelligence
topic modeling, online reputation, social media analysis, trust, artificial intelligence
1. Hattie J. Visible Learning: a Synthesis of Over 800 Meta-Analyses Relating to Achievement. London: Routledge, 2011.
2. A Topic Modeling Comparison Between LDA, NMF, Top2Vec, and BERTopic to Demystify Twitter Posts.
3. Pavithra & Savitha. Topic Modeling for Evolving Textual Data Using LDA, HDP, NMF, BERTOPIC, and DTM With a Focus on Research Papers // Journal of Technology and Informatics (JoTI). 2024. No. 5. Pp. 53–63. DOI:https://doi.org/10.37802/joti.v5i2.618
4. SocialVisTUM: An Interactive Visualization Toolkit for Correlated Neural Topic Models on Social Media Opinion Mining / G. Hagerer [et al.] // Proceedings of the International Conference on Recent Advance. 2021.
