Instruments and Systems: Monitoring, Control, and Diagnostics Annotation << Back Study of the Digital Shadow of Russian NPP Construction Projects Abroad Using Text Mining Nethods A.I. GUSEVA, V.S. KIREEV, P.V. BOCHKARYOV This article is devoted to the study of the level of public acceptability of nuclear power in three countries where Russian NPPs are being built – the Republic of Belarus, Hungary and Turkey. Public opinion, reflecting the level of acceptability, is formed from such components as the level of knowledge about the nuclear industry, public awareness of the economic and social benefits from nuclear energy and consumer installations. To analyze the consumer’s installations, the data accumulated in the digital shadow of NPP construction projects (on Internet servers, electronic media, social networks, etc.) for the period 2019–2022 were taken. The Internet activity of users in relation to the topic "nuclear energy", the tone of text publications in electronic media and social media videos, the presence of opinion leaders and influencers and the tone of their statements were analyzed. Sentiment analysis was based on intelligent text analysis methods based on machine learning and CNN convolutional neural networks. A graph approach was used to identify influencers. As a result, for the three countries under study, the level of social acceptability was determined by the author’s method as quite high (0.65–0.78). This is confirmed by the fact that, despite the strongest external pressure and a large number of anti-Russian sanctions, the governments of the studied countries continue to implement construction projects without changing the original plan. Keywords: digital shadow, intelligent text analysis methods, convolutional neural networks, ensemble clustering, graph method for identifying influencers. DOI: 10.25791/pribor.6.2023.1417 Pp. 50-58.