Special Issue 'Online Fuzzy Machine Learning and Data Mining' (Information Sciences)
Sprache des Titels:
Englisch
Original Kurzfassung:
This special issue intends to investigate the relationship between fuzzy set theory and
ML/DM with special emphasis on (but not restricted to) a particular class of approaches
within the field of Fuzzy ML-DM dealing with on-line, incremental learning methods. The
aim is to investigate incremental adaptation of the model parameters and the evolution
of the model as cornerstone elements of techniques dedicated to dynamically changing
environments over time and space. Typically, data streaming exemplifies dynamic systems
(with changing operation conditions and system characteristics) which can be found in
various industrial and rich-data applications (e.g. control, robotics, web, etc.). Fuzzy learning
models for such systems depart from the idea that memory cannot suffice to handle all data
in a one-shot experiment (e.g. in the case of huge data bases or web applications). Data is
therefore segmented and processed sequentially and incrementally in an online way. In pure
online applications, individual data samples arrive over time requiring again incremental
processing. This special issue intends to draw a picture of the recent advances in fuzzy
online learning as a bridge between online ML and DM on one side and fuzzy theory
on the other side.