Marek Reformat

Area of Study / Keywords

Software Engineering and Intelligent Systems Computer Engineering Signal and Image Processing Energy Systems

About

Marek Reformat received his MSc degree (with honors) from Technical University of Poznan, Poland, and a PhD degree from University of Manitoba, Canada.

His initial research projects involved different aspects related to computer networks, especially in the area of management and performance measurement. He co-authored several papers and reports regarding this topic. During his PhD studies, his research interests included distributed computing, with emphasis on fault-tolerant systems in such frameworks as Parallel Virtual Machine (PVM) and Message Passing Interface (MPI); optimization methods; and fuzzy sets and systems. His principle interest was related to evolutionary computing and its application to optimization problems. He proposed a new methodology for design of control systems, which relied on a combination of advanced system simulators and genetic computation. He applied this concept to the control design problem in the area of power systems. In 1997 he joined the Manitoba HVDC Research Centre, where he was a member of a simulation software development team. He was involved in improvement and development of an electromagnetic transients program for time-domain simulation, performed functional and structural testing of the software, and provided expert consulting services in the area of simulation and modeling internationally.

Marek has been with the Department of Electrical and Computer Engineering at University of Alberta since July 2000. He is Professor and Associate Chair of Graduate Studies in the Department. In addition, he is an Associate Editor of a number of journals related to computational intelligence and software engineering. He has been a member of program committees of several conferences related to those areas. He is actively involved in North American Fuzzy Information Processing Society (NAFIPS). He is a member of the IEEE and ACM.

Research

Research Interests

The goal of Marek Reformat’s research activities is to develop methods and techniques for modeling data and knowledge, as well as design systems that possess abilities to imitate different aspects of human behavior. In this context, he uses concepts of computational intelligence — granular (fuzzy) computing, neuro computing, and evolutionary computing — as key elements necessary for capturing relationships between pieces of data and knowledge, as well as for mimicking human ways of reasoning about opinions and facts. He combines these methods with techniques capable of dealing with uncertainty - Bayesian systems, and Dempster-Shafer's evidence theory. These activities focus on introduction of human aspects to software systems, and development of more human-aware and human-like systems.

Current Research

The current research projects embrace the following areas:

• Knowledge extraction and knowledge representation: application of fuzzy, neurofuzzy and evolutionary computing methods to discovery and representation of knowledge.
• Decision support: application of different forms of knowledge representation to construction of systems supporting decision-making processes; integration of uncertainty — expressed using probability theory (Bayesian networks), Dempster-Shafer theory, and fuzzy measures — with decision support systems.
• Semantic-based intelligent systems: development of intelligent systems capable of providing more human-like outcomes based on ontologies and reasoning engines; applications include computing with words, semantic web, and systems for information analysis.
• Software quality and maintenance: development of models for estimation and prediction of quality and maintenance related aspects of software components.

ECE 220 - Programming for Electrical Engineering

Architecture and basic components of computing systems. Programming environment and program development methodology. Basics of programming: from data structures and functions to communication with external devices. Principles of object-oriented programming. Good programming style. Prerequisite: ENCMP 100.

ECE 447 - Data Analysis and Machine Learning for Engineers

The course introduces basic concepts and techniques of data analysis and machine learning. Topics include: data preprocessing techniques, decision trees, nearest neighbor algorithms, linear and logistic regressions, clustering, dimensionality reduction, model evaluation, deployment methods, and emerging topics. Prerequisites: ECE 220 or CMPUT 275, and ECE 342 or STAT 235, or consent of instructor.

ECE 720 - Advanced Topics in Software Engineering and Intelligent Systems

Browse more courses taught by Marek Reformat