AI4TI Laboratory

The laboratory develops information systems: from architecture to working versions.

Table of contents:

• Information
• Projects
• Research
• Other achievements

Information

List of development directions:

• Development of information systems architecture
• Web application development
• Development of mathematical models
• Development of machine learning models
• Conducting scientific research

Projects

	PalMon A project in which a hardware and software system was developed to determine damage to cargo on a pallet. The project developed: 1. Information architecture systems 2. Web application for viewing damage information 3. Mathematical model for determining cargo damage 4. Hardware for reading data from a pallet
	MacCheck A project in which a web application has been developed to manage network equipment through a user interface. This information system helps system administrators more easily manage and control equipment in the enterprise.
	PowerGuard A project to identify faults in an electrical circuit signal based on mathematical models and methods. The project implements a web application and a mathematical module that localizes and classifies failures based on the Wavelet transform, optimization methods, parallel algorithms and other mathematical methods.
	Emotional View A project to determine the level of client satisfaction based on the dynamics of the client's emotions. The project used machine learning methods and neural networks. The project includes a hardware and software system for reading time series data on the dynamics of emotions and subsequent prediction of the level of satisfaction.
	Pichta A project in which a web application was implemented to compile a graph of competencies for a user-selected specialization. The system uses up-to-date data on competencies indicated in the latest vacancies of various companies.

Research

MAMGD Optimizer Gradient optimization method using exponential damping and second-order discrete derivative for neural networks and multidimensional real functions About

WONC-FD Method WONC-FD (Wavelet-Based Optimization and Numerical Computing for Fault Detection) is a method for detecting and classifying faults in time-series data using wavelet analysis and numerical optimization. This project aims to provide a robust and efficient method for identifying various types of errors within a signal. About

DMD-Neural-Operator Solving partial differential equations (PDEs) for various initial and boundary conditions requires significant computational resources. We propose a neural operator $G_\theta: \mathcal{A} \to \mathcal{U}$, mapping functional spaces, which combines dynamic mode decomposition (DMD) and deep learning for efficient modeling of spatiotemporal processes. The method automatically extracts key modes and system dynamics and uses them to construct predictions, reducing computational costs compared to traditional methods (FEM, FDM, FVM). The approach is demonstrated on the heat equation, Laplace's equation, and Burgers' equation, where it achieves high solution reconstruction accuracy. About

Other achievements

QR-Detect QR_Detect is a web application for detecting and recognizing QR codes based on video recordings. QR_Detect uses machine learning techniques to localize and translate QR codes into URLs. Data on recognized codes is recorded in a database with additional information.