An overview on evolving systems and learning from stream data

Abstract

Evolving systems unfolds from the interaction and cooperation between systems with adaptive structures, and recursive methods of machine learning. They construct models and derive decision patterns from stream data produced by dynamically changing environments. Different components that assemble the system structure can be chosen, being rules, trees, neurons, and nodes of graphs amongst the most prominent. Evolving systems relate mainly with time-varying environments, and processing of nonstationary data using computationally efficient recursive algorithms. They are particularly appropriate for online, real-time applications, and dynamically changing situations or operating conditions. This paper gives an overview of evolving systems with focus on system components, learning algorithms, and application examples. The purpose is to introduce the main ideas and some state-of-the-art methods of the area as well as to guide the reader to the essential literature, main methodological frameworks, and their foundations.