Перегляд за Автор "Huskova, V. H."

Зараз показуємо 1 - 3 з 3
  • Ескіз
    ДокументВідкритий доступ
    Artificial intelligence-based recommendation system using cloud technologies
    (Igor Sikorsky Kyiv Polytechnic Institute, 2024) Sulima, Ivan; Huskova, V. H.
    Bachelor thesis: 85 p., 20 figures, 6 tables, 40 references, 1 appendix. The object of research is development of architecture on top of the cloud that uses recommendation mechanisms. The subject of research are a methods and technologies that are used in cloud architection and recommender systems. The aim of the work is to build a system (architecture) using cloud technologies, in the center of which will be a recommendation system. The relevance of this work is associated with possibility of great impact on products revenue with help of personalized recommendations. The research keeps attention on the challenge of developing a recommendation system and predicting user preferences, with a specific focus on utilizing the Cloud technologies. This involves justifying the use of particular methods, tools, and the Cloud development environments, preparing and analyzing the initial dataset, preprocessing raw data, and developing mathematical models tailored to the recommendation problem. The final solution is implemented on AWS using Personalize and then the results are analyzed. Experiments and comparisons with existing approaches are conducted to evaluate the effectiveness of the recommendation system developed on AWS. The experimental results demonstrate the advantages of the Cloud deployment and Recommender systems in general.
  • Ескіз
    ДокументВідкритий доступ
    Development of a personalized content recommendation system for streaming platform based on user behavior analysis
    (Igor Sikorsky Kyiv Polytechnic Institute, 2024) Aliiev, Ramin Nazirovich; Huskova, V. H.
    Thesis: 101 pages, 18 tables, 26 figures, 1 appendix, 27 sources. The object of research is the process of creating a recommendation system for selecting movies on a streaming platform. The subject of research is the methods and algorithms used to create content personalization systems. The aim of the work is to investigate methods and algorithms for building recommendations and increasing their efficiency, and to integrate the selected algorithm into a movie recommendation application. The relevance of this work is associated with the need to predict user preferences based on data collected from their interactions and feedback. The research involves analyzing the problem of developing a recommendation system and predicting user preferences, justifying the use of specific methods, tools, and development environments, preparing and analyzing the initial dataset, preprocessing raw data, developing mathematical models for the recommendation problem, implementing the solution, and analyzing the results obtained. Experiments and comparisons with existing approaches are conducted to evaluate the effectiveness of the developed recommendation system. The results of the experiments demonstrate the advantages of the developed movie recommendation system for the streaming platform.
  • Ескіз
    ДокументВідкритий доступ
    Time series analysis and forecasting of demographics of developed and developing countries
    (Igor Sikorsky Kyiv Polytechnic Institute, 2024) Maznichenko, Lev Vladyslavovych; Huskova, V. H.
    Bachelor thesis: 107 p., 6 figures, 5 tables, 32 references, 3 appendixes. The object of the study is the demographics of developed and developing countries. The subject of research is the time series analysis and forecasting methods applied to demographic data. The purpose of the work is to is to develop and evaluate a model for predicting demographic trends in various countries using time series analysis techniques. The relevance of this thesis lies in the development of the field of demographics (TFR) forecasting, which is essential for understanding and planning for the needs of society and Humankind itself. Accurate population projections help in policy making, resource allocation and economic planning. Traditional models often fail to capture the nuances of demographic changes, especially under different economic, political and cultural conditions. During research, a sophisticated forecasting model was developed to predict demographic trends in both developed and developing countries. This model uses advanced time series analysis and forecasting techniques to provide increased accuracy and insight into future demographic shifts, thereby assisting in effective decision-making and planning.