Introduction
Evaluation of ARTMAP classifiers - Dependence on cluster representation Large number of classification algorithms available without detailed knowledge of their properties/performance. Neural networks are considered to be assumption-less classifiers (do not require explicit assumptions about data distribution). But these systems have implicit assumptions built into them. These assumptions are related to the data representation and algorithmic properties of the system. In this study, the dependence of the ARTMAP classifiers performance on the internal cluster representation is analyzed for image data from remote sensing.
Data Set Seven-dimensional Landsat TM image of the city of Košice (Figure 1) Size of image: 368,125 pixels, out of which 6,331 classified by an expert into seven categories (A - urban area, B - barren fields, C - bushes, D - agricultural fields, E - Meadows, F - Woods, and G - water) Method of analysis The performance is compared in terms of the weighted PCC (Percent of Correctly Classified) and the contingency tables. Compared systems Fuzzy ARTMAP, Gaussian ARTMAP, Extended Gaussian ARTMAP.
Figure 1: Original image
ARTMAP classifier topology
Confidence index
Rock quarry identification (Oto)
Results
Other tested modifications - results???
Study of voting ??? (not probable -- I can make it)
Conclusions
References