The goal of this paper is to outline the author’s perception of current research in fuzzy machine learning, which includes the discussion of the role of fuzzy sets in machine learning. This perception is based on significant experience with both research communities, fuzzy logic and machine learning, not only as an author of research papers but also as a reviewer, conference organizer and journal editor. In spite of this, it goes without saying that the presentation will necessarily remain subjective and potentially biased.
Prior to proceeding, it should be emphasized that this is not a survey paper. In fact, references are rather sparse and sometimes deliberately omitted (especially in connection with more critical comments or negative remarks), and the fraction of self-citations is higher than usual. Moreover, the focus of this paper is more on machine learning (model induction) and less on data mining (pattern mining, exploratory data analysis, data description). Although both fields are closely connected, there are nevertheless important differences between them, and these differences are not unimportant with regard to the possible role and potential contributions of fuzzy logic—see [16]for a more detailed discussion of this point.