Search academic texts

Information × Registration Number 0224U032663, (0123U102677) , R & D reports Title Methods and means of research of markers of aging and their influence on postcovid effects for prolongation of the working period popup.stage_title Розроблення методу пошуку залежностей у великих наборах. Розроблення моделі зміни маркерів старіння в регіоні Head Shakhovska Nataliia B., Доктор технічних наук Registration Date 29-11-2024 Organization Lviv Polytechnic National University popup.description1 The aim of the project is to determine the markers of aging based on big data technologies and methods of analysis of small data samples and to build models of biological age assessment to assess the impact of COVID19 and prolong healthy life expectancy and efficiency. popup.description2  In the framework of the scientific project executed in 2024, significant advancements were achieved in addressing pressing data analysis challenges. A novel approach for imputing missing data in big data interfaces was developed, substantially improving analysis quality under conditions of uncertainty. An intra-ensemble model for intelligent analysis of short medical data sets was implemented, offering enhanced accuracy and better generalization. Additionally, the research substantiated hidden data dependencies, optimizing decision-making processes. A dependency search method for large data sets with gaps was introduced, based on a combination of clustering and auto-associative methods, which reduces computational complexity. The architecture of Decision Support System for Navigation in Environments with Incomplete Information was created, capable of handling uncertainties of up to 40%. A significant achievement was the development of the SEIR(D) geospatial model for simulating the spread of infectious diseases. This model incorporates behavioral and geospatial data to predict pandemic parameters, including the number of infected, recovered, and deceased individuals. The model was validated using COVID-19 data from Lviv, Ukraine. Moreover, a big data analysis method was designed to generate advisory rules supporting governmental policy formulation. An in-depth analysis of cloud data storage architectures was conducted, resulting in the proposal of a new virtual storage architecture supporting NoSQL data (InfluxDB, Neo4j, MongoDB). This approach significantly reduces capacity complexity and enhances the efficiency of recursive query execution. Product Description popup.authors Izonin Ivan V. Boiko Nataliia I. Vykliuk Yaroslav I. Kaminskyi Roman M. Kryvenchuk Yurii P. Melnykova Nataliia I. Tkachenko Roman O. Yakovyna Vitalii S. popup.nrat_date 2024-11-29 Close