Search academic texts

Information × Registration Number 0225U002750, (0123U101687) , R & D reports Title Methods and means of information security and hygiene based on explainable artificial intelligence popup.stage_title Розроблення методів аналізу безпечності ПЗ та моделювання поширення фейкових новин Head Shakhovska Nataliia B., Доктор технічних наук Registration Date 01-04-2025 Organization Lviv Polytechnic National University popup.description1 The goal of the project is to increase the transparency and interpretability of the results of machine learning models in the field of information security and hygiene popup.description2 As a result of the implementation of the 2nd stage of the project, a method for analyzing the level of defect resistance of a real software product was developed; a method for two-stage processing of data of different volumes based on a hierarchical classifier; a method for analyzing hidden dependencies based on the use of supervised and unsupervised machine learning models, which formed the basis of a new neural network architecture with a fractal structure. A data preparation algorithm was developed, a method for clustering different types of data was developed, which allows working with streaming data based on division into packages, an analysis and comparison of methods for processing missing values was conducted, and a statistical method of percentiles for calculating initial centroids was analyzed.  Product Description popup.authors Yemets Kyrylo V. Izonin Ivan V. Vovk Olena B. Markelov Oleksandr E. Mykhailyshyn Vladyslav Yu. Serov Yurii O. Tkachenko Roman O. Fedushko Solomiia S. Shymanskyi Volodymyr M. Yakovyna Vitalii S. popup.nrat_date 2025-04-01 Close