Work place: Institute of Computer Science and Information Technologies, Lviv Polytechnic National University, Lviv, Ukraine
Research Interests: Data Structures and Algorithms, Image Processing, Neural Networks, Machine Learning, Artificial Intelligence
Izonin Ivan Viktorovych is an Associate Professor at the Department of Artificial Intelligence of Lviv Polytechnic National University, Ukraine. I received my MSc degree in Computer science at Lviv Polytechnic National University in 2011 and MSc degree in Economic cybernetics at Ivan Franko National University of Lviv in 2012. I received a Ph.D. in Artificial Intelligence at Lviv Polytechnic National University in 2016. My main research interests are focused on the high-speed computational intelligence, neural-like structures, non-iterative training algorithms, ensemble models, meta learning and small data analysis.
He is currently working at the High-speed Computational Intelligence Lab. Our Lab is focusing on information modeling based on a new ANN design concept - Successive Geometric Transformations Model (SGTM). SGTM ensures the solutions to a lot of tasks (pattern recognition, predicting, classification, optimization, missing-data recovering or their consolidation, information protection and privacy methods...). Neural-like structures based on the SGTM as universal approximators implement the training and self-training principles and are based on algorithmic or hardware performing variants. SGTM uses a single methodological framework for various tasks and fast non-iterative training with the pre-defined number of computation steps. It provides repeatability of the training outcomes and the possibility to obtain satisfactory solutions for large and small training samples.
DOI: https://doi.org/10.5815/ijisa.2018.09.05, Pub. Date: 8 Sep. 2018
This document presents two developed methods for solving the classification task of medical implant materials based on the compatible use of the Wiener Polynomial and SVM. The high accuracy of the proposed methodology for solving this task are experimentally confirmed. A comparison of the proposed methods with existing ones: Logistic Regression; Linear SVC; Random Forest; SVC (linear kernel); SVC (RBF kernel); Random Forest + Wiener Polynomial is carried out. The duration of training of all methods that described in work is investigated. The article presents the visualization of all method results for solving this task.[...] Read more.
DOI: https://doi.org/10.5815/ijisa.2016.12.01, Pub. Date: 8 Dec. 2016
The paper describes the image superresolution method with aggregate divergence matrix and automatic detection of crossover. Formulation of the problem, building extreme optimization task and its solution for solving the automation determination of the crossover coefficient is presented. Different ways for building oversampling images algorithms based on the proposed method are shows. Based on practical experiments shows the effectiveness of the procedure of automatically the determination of the crossover coefficient. Experimentally established the effectiveness of the procedures oversampling images at high zoom resolution by the developed method.[...] Read more.
Subscribe to receive issue release notifications and newsletters from MECS Press journals