The research programme of the CoE UNITe is focused on the following main topics:
- Innovative mathematical methods and models to support digital transformation
The main challenge to be solved is the development, evaluation and implementation of reliable methods for data discovery and assessment, secure data storage, efficient data processing, statistical data analysis and multivariate data visualization, that can be integrated into a single workflow and thus be applicable in the creation of intelligent software tools supporting informed decision-making in various areas of public importance. An important goal is the development of generic software tools and solutions for a broad range of mathematical problems with applications in natural science, humanities, education, technology and industry.
Main tasks:
- Advanced methods providing quality of data
- Advanced solutions in data coding and security
- New statistical methods for data analysis, applicable in biology, cancer research, demography
- New algorithms for modelling and visualization of geometric objects
- Development of generic mathematical software
- Data-Centric AI Applications
The main objective is to explore and solve challenges in the field of Deep Learning (DL), the limitations of Large Language Models (LLM), the causes of inappropriate or harmful content generation, the mechanisms behind AI predictions, and the need for significant resources for training, updating, and use.
Main tasks:
- Developing new and experimenting with existing approaches for the integration of Deep Learning with classical symbolic-based approaches to improve the performance and explainability of the models;
- Development and adoption of tailored models for managing the resource limitations and energy efficiency of the models;
- Development of specific custom AI architectures that can deal with multimodal information.
- Development of practical AI-empowered solutions for socially significant problems such as: how to deal with disinformation; practical methods for analysis of text, images, and data for improved healthcare and/ digitalisation of medical and cultural resources; application of AI approaches for solving emerging engineering problems such as green and sustainable technology development.
- Development of unified AI architectures and frameworks including various methods, models and tools, and pilot application in National Library, Plovdiv (text processing), Bulgarian National Radio (audio processing), Troyan History Museum (image processing).
- Innovative methods, technologies and tools for modern technology-enhanced learning
The main goal is to explore how adaptive and personalized learning systems can adjust to the pace, style, and progress of learners, providing individualized learning, and also how technologies such as VR and AR can be effectively integrated into TEL to create an interactive and engaging learning environment.
Main tasks:
- Experimental development of a technology framework for educational video games with enhanced learnability and playability
- Novel model, scenarios, and tools for interactive and gamified learning activities for an increased engagement and motivation of learners and collaboration among students and educators
- Design and technology validation of two adaptive and personalized educational video games using the technology framework
- Practical validation of an interactive learning environment by integration of VR, AR, and educational video games, with learning analytics of personalized learning strategies and continuous improvement in an immersive TEL
- Validation of an innovative model, methods, and tools for enhancing security in TEL
- Technology pilot demonstrations in educational institutions – schools and universities
- Bioinformatics and crisis prevention
The main objective is developing effective and efficient methods and tools for processing large volumes of data generated by sequencers applying powerful computing resources and efficient big data processing algorithms. Special focus is given on integrating genomic sequences, experimental results, protein classifications, and other types of data used in bioinformatics analyses, into practical and useful medical applications. Other group of challenges is related to developing sufficiently accurate, fast, and optimal mathematical models of real processes, and for crisis management for natural disasters and environmental control.
Main tasks:
- Testing various approaches for processing large volumes of data generated by sequencers and finding ways to measure and store them
- Optimizing and improving existing approaches to integrate various types of data used in bioinformatics analyses, such as genomic sequences, experimental results, protein classifications, mutation identification, and more
- Developing or improving applications of methods for analyzing genomic variations, discovering their functionsand structural variations, to improve the understanding of the genetic bases of various diseases and phenomena
- Developing and refining approaches, methods, and algorithms for creating intelligent models for analysing real processes
- Creating methods, algorithms, models, and prototypes for describing and analysing crisis management mechanisms for natural disasters and environmental control
- Creating methods for modelling and predicting the frequency of occurrence of dangerous and anomalous temperatures and the occurrence of hazardous natural disasters
- 3D technologies for innovative and sustainable future
The 3D technologies are facing many challenges. The majority of the present digitalization technologies are only able to create 3D models of static objects, which is hindering the digitalization of processes and movements, like national dances (intangible heritage), sport movements, etc. The real-time 3D capture of movements has also huge application in the medical sector, where it can be used to monitor the progress of different treatments for motor function diseases.
Main tasks:
- Analytical and experimental modelling and evaluation of various 3D technologies and processes;
- Business modelling in the area of 3D technologies;
- Studies and experiments on the applications of 3D technologies in the material science and the manufacturing processes;
- Development, validation and practical demonstration of approaches, methods and solutions for digitization and visualization of heritage;
- Development of innovative methods and improvement of the existing solutions for application of 3D technologies in the creative and recreative industries, as well as in the medical and healthcare sectors;
- Development of methods and approaches for use of 3D technologies for surveying purposes and in the forensic sciences;
- Smart and Sustainable Systems
The scientific activity is aimed at research and innovation in the field of cyber-physical systems, digital twins and sustainable technologies, with a main focus on their application in industry, energy management and smart cities.
Main tasks:
- Optimization of the manufacturing processes – methods for data and information transfer, allowing for the automation and integration of corporate processes in order to attain best practices in process management.
- Digital twin for smart manufacturing – create a virtual simulation of a physical system and process for identify potential problems and opportunities for improvement.
- Digital twin for smart and sustainable cities – design new approaches that support a digital twin in smart cities based on the data lifecycle management process and used to discovery possible problems in city.
- Data interoperability for smart and sustainable systems – exchange contextualized information to better operational decisions and prediction outcomes.
- Cybersecurity: vulnerability analysis, prevention and protection from cyber attacks
The activity is focused on developing advanced solutions to protect digital systems, networks and infrastructures from growing cyber threats, with the main emphasis on intelligent methods that use artificial intelligence, machine learning and the Internet of Things (IoT) to detect, prevent and respond to attacks.
Main tasks:
- Assessment of threats and risks, controlling access according to the needs of the end users;
- Analysis of cyberattacks with a large defeat on communication and information systems and to propose technical measures to ensure cyber security;
- Detecting vulnerabilities of software products and network structures;
- Determining exploits and developing an algorithm to prevent and combat cyber-attacks;
- Protecting critical infrastructure, increasing autonomy, development of a Cyber Attack Autonomy Model.