Evolving Fuzzy Systems (EFS) is a new and dynamic branch of fuzzy systems theory and applications has emerged as a result of the efforts to address some new challenges. The problem of automatic design of fuzzy systems for modeling, classification, time-series prediction, regression, clustering etc. from data has been successfully addressed in the off-line case by a range of techniques such as gradient-based (neuro-fuzzy approach, ANFIS), genetic or more generally evolutionary algorithms based (nowadays well established GFS), using partitioning by clustering or experts and learning by least squares techniques etc. A range of new challenges appear during the last decade or so, which require completely new approaches. We are in the midst of an information revolution witnessing an exponential growth of the quantity and the rate of appearance of new information by; Internet users, consumers, finance industry, sensors in advanced industrial processes, autonomous systems, space- and aircrafts etc. The new challenges that cannot be successfully addressed by the existing techniques can be summarized as follows: i) to cope with huge amounts of data; ii) to process streaming data on-line and in real-time; iii) to adapt to the changing environment and data pattern by systems (models, classifiers, predictors etc. that have flexible, open, expandable, as we say, ‘evolving’ structure); iv) to be computationally efficient (that means to use recursive, one-pass, non-iterative approaches); to preserve the interpretability and transparency in a dynamic sense. We need efficient approaches to deal with data streams, not just with batch sets of data, to detect, react and take advantage of concept shift and drift in the data streams. We need efficient collaborative and interactive schemes for a range of applications in process industry (for self-calibrating, self-maintaining intelligent sensors of new generation), in autonomous systems and robotics (for systems that have self-awareness, re-planning and knowledge accumulation capabilities), in multimedia and biomedical applications to name the few.

The research topic of evolving fuzzy systems (EFS) emerged during the last 5 years or so as a collective effort of a relatively small number of active researchers. It has now been recognized through a range of activities, such as: i) IEEE International Symposium on Evolving Fuzzy Systems 2006 (Proceedings published by IEEE Press); ii) a special issue of IEEE Transactions on Fuzzy Systems, 2007 with over 20 papers received; iii) an edited book by John Wiley, 2008; iv) a large number of special sessions and Tutorials at leading IEEE Conferences; iv) a growing number of publications, patents, and industrial applications in this emerging area of research.