Evolving intelligent systems (EIS) are highly adaptive systems able to update its own parameters and structure based on a date stream. These systems have been developed to address problems of modeling, control, prediction, classification and data processing in a nonstationary, dynamic changing environment. Pioneers works in this area are dated from the around the turn of the centuries and were focused in areas of neural networks, fuzzy rule-based systems and neural-fuzzy hybrids. In this century the area has been expanded to also address statistical models, hardware implementations and so on. The aim of this chapter is to provide an introduction and a state of the art view about this subject. The purpose is to present the paradigm and the associated concepts, address the main learning approaches, and detail recently developed models based on participatory learning and fuzzy trees. © Springer-Verlag Berlin Heidelberg 2013.
CITATION STYLE
Lemos, A., Caminhas, W., & Gomide, F. (2013). Evolving intelligent systems: Methods, algorithms and applications. Smart Innovation, Systems and Technologies, 13, 117–159. https://doi.org/10.1007/978-3-642-28699-5_6
Mendeley helps you to discover research relevant for your work.