Search academic texts

Information × Registration Number 0226U001655, (0125U001683) , R & D reports Title Methodology for determining tonality and classification of multimodal content in territorial revitalization projects based on neural network methods popup.stage_title Створення детального плану проєкту, включно з розподілом відповідальності та ресурсів Head Honcharenko Tetiana A., Доктор технічних наук Registration Date 29-01-2026 Organization Kyiv National University of Construction and Architecture popup.description1 The purpose of the study is to solve the theoretical, methodological and scientific-applied problem associated with the formation of a single information environment and analytical and digital space for information support and formalized adjustment of decisions of investment and construction projects at the stage of spatial planning and development of territories through the development of information technologies and the involvement of an interdisciplinary approach to the processes of aggregation and generalization of territorially distributed spatial information from various subjects of the formation of information resources. The results of the study are aimed at creating an information technology for processing spatial and attributive information from various state registers, sectoral cadastral and information services, ensuring interdisciplinary integration of data from various sources, for the formation of complete information about spatial planning objects and their presentation in digital cartographic form for the automated development of planning project solutions. popup.description2 The object of the study is the processes of automated processing of multimodal information in territorial revitalization projects. The subject of the study is the methodology, models, methods and information technology of developing a unified information environment for automated processing of multimodal information based on neural network methods. The purpose of the study is to solve the theoretical, methodological and scientific and applied problem associated with the formation of a unified information environment and analytical and digital space for information support and formalized adjustment of decisions of investment and construction projects at the stage of spatial planning and development of territories through the development of information technologies and the involvement of an interdisciplinary approach to the processes of aggregation and generalization of territorially distributed spatial information from various subjects of information resource formation. As a result of the first stage of the research, a detailed project plan was created, including the distribution of responsibilities and resources, and a theoretical basis was studied for developing methodological foundations for the application of neural network methods to form a single information environment for automated processing of multimodal information in territorial revitalization projects, on the basis of which the concept of explanatory artificial intelligence for multimodal sentiment analysis will be developed, specially adapted to the requirements of managing territorial revitalization projects. Keywords: GAN technology, methods of optimization and adaptation of neural networks, convolutional neural network, ontology, methods of pre-processing images, artificial intelligence, methods of intelligent data analysis, BIM model, territorial revitalization. Product Description popup.authors Oleksandr S. Molodid Olha L. Solovei Yuliia Riabchun Oleksandr A. Poplavskyi Maksym Delembovskyi Serhii Dolhopolov popup.nrat_date 2026-01-29 Close