Generalized nets model of dimensionality reduction in time series

Abstract

The paper considers the generalized nets as an extension of Petri nets applied for modeling of the methodology called Symbolic Essential Attributes Approximation (Krawczak and Szkatuła, 2014). SEAA was developed to reduce the dimensionality of multidimensional time series by generating a new nominal representation of the original data series. In general the approach is based on the concept of data series envelopes and essential attributes obtained by a multilayer neural network. The symbolic data series representation -which just describes the compressed representation of the original data series -is obtained via discretization of the real-valued essential attributes. In this paper the generalized nets were used to model the logistic of processes involved in SEAA methodology. First the basic of the theory of generalized nets is introduced, next SEAA methodology processes are modeled via the generalized nets the new model of SEAA.