The 12th International Conference on Fuzzy Systems and Data Mining (FSDM 2026)

Abstract: We will present special techniques of fuzzy modeling suitable for applications in time series processing, namely the Fuzzy Transform (F-transform) and selected methods of Fuzzy Natural Logic (FNL). The F-transform is applied to estimation of the trend or trend-cycle of time series, and to estimation of the slope of time series over an imprecisely specified area. Our methods are based on the decomposition of the time series into 4 components: trend, cycle, seasonal component, and random disturbances. The fuzzy transform makes it possible to find arbitrary shape of the trend or trend-cycle. It has been proved that using the F-transform, we can eliminate seasonal component and significantly reduce noise. Moreover, the computational complexity is low.
Methods based on FNL have applications in mining information from time series. Among them, let us mention reduction of dimensionality, finding intervals of monotonous behavior and their characterization using expressions of natural language, finding perceptually important points, measure of similarity between time series, or automatic summarization of knowledge about time series. We also suggest a powerful method for detection and statistical testing of structural breaks in time series.
We will compare our methods with traditional methods and demonstrate that they can on one hand successfully compete with them and on the other hand, both kinds of methods can be combined to increase the effectivity of the processing of time series.

Biography: Prof. Vilém Novák, DSc. is founder of the Institute for Research and Applications of Fuzzy Modeling of the University of Ostrava, Czech Republic. He obtained PhD in mathematical logic at Charles University in Prague, DSc. (Doctor of Sciences) in computer science in Polish Academy of Sciences, Warsaw, full professor at Masaryk University, Brno. His research activities include mathematical fuzzy logic, approximate reasoning, mathematical modeling of linguistic semantics, fuzzy control, processing of time series, and applications of fuzzy modeling. He is pioneer of the fuzzy set theory and author or co-author of 6 monographs and over 350 papers with almost 10000 citations.