Frobenius Perron Graphs for Temporal Data Classification

B.J. van Wyk and M.A. van Wyk (South Africa)

Keywords

Temporal Reasoning, Modelbased reasoning, Self Or ganizing Feature Map, Graphical Models, Time Series Classification.

Abstract

In this paper Frobenius Perron Graphs (FPGs), based on the discrete Frobenius Perron operator, are introduced. An algorithm for the derivation of Frobenius Perron Graphs (FPGs) and a method to use FPGs for the classification of time series data are presented. It is shown that one can construct an FPG for each class of time series data and then use an Edge Accumulator (EA) for classification. FPG classification is applied to the classification of time series data and simulation results indicate that the FPG method is not only robust, but also easy to implement.

Important Links:



Go Back