Agnieszka Jastrzębska, PhD

The purpose of the research was to investigate and supplement existing approach to time series modelling using the Fuzzy Cognitive Maps, which is a powerful concepts-based modelling framework suitable to represent complex phenomena.

A broad survey of techniques for time series modelling with Fuzzy Cognitive Maps has been carried out. Several missing elements have been identified and some new solutions have been propsoed. The theoretical study was accompanied by the empirical research and both served as a thorough demonstration of superior modelling properties of Fuzzy Cognitive Maps.

Under investigation were properties of time series representation for modelling with Fuzzy Cognitive Maps. A new representation has been introduced and compared with an approach known in the literature. It has been also demonstrated that Fuzzy Cognitive Map design strategy heavily influences modelling outcome and a new method for Fuzzy Cognitive Map design based on cluster validity indices has been introduced. 

All results are documented in a number of papers, a list of which may be find  here.

You are here: Home Results Agnieszka Jastrzębska, PhD