Visualization of Fuzzy Rule Classifiers for Flight Duration Forecast

The impact of the weather on the flight duration of aircraft has been analysed in various studies. The complex aspects of the weather, which are accordingly reflected in the weather data, demand sophisticated techniques to visualize the analytical results. In this paper we present an approach to visualize fuzzy rules describing high-dimensional data. By means of this method, the rules, as well as the classified data, can be presented on an arbitrary low-dimensional space. We will demonstrate the efficiency of this technique on some benchmark examples and on real weather data set that is used to predict aircraft flight duration on a European hub airport.