|
|
BINARY CLASSIFICATION STRATEGIES FOR MAPPING URBAN LAND COVER WITH ENSEMBLE CLASSIFIERS Host Publication: Finds and Results from the Swedish Cyprus Expedition: A Gender Perspective at the Medelhavsmuseet Authors: J. C-W Chan, L. Demarchi, T. Van De Voorde and F. Canters Publication Date: Aug. 2008 Number of Pages: 4
Abstract: We investigated two binary classification strategies to further extend the strength of ensemble classifiers for mapping of urban objects. The first strategy was a one-against-one approach. The idea behind it was to employ a pairwise binary classification where n(nǃ)/2 classifiers are created, n being the number of classes. Each of the n(nǃ)/2 classifiers was trained using only training cases from two classes at a time. The ensemble was then combined by majority voting. The second strategy was a one-against-all binary approach: if there are n classes, with a = {1,..., n} being one of the classes, then n classifiers were generated, each representing a binary classification of a and non-a. The ensemble was combined using accuracy estimates obtained for each class. Both binary strategies were applied on two single classifiers (decision trees and artificial neural network) and two ensemble classifiers (Random Forest and Adaboost). Two multi-source data sets were used: one was prepared for an object-based classification and one for a conventional pixel-based approach. Our results indicate that ensemble classifiers generate significantly higher accuracies than a single classifier. Compared to a single C5.0 tree, Random Forest and Adaboost increased the accuracy by 2 to 12%. The range of increase depends on the data set that was used. Applying binary classification strategies often increases accuracy, but only marginally (between 1Dž%). All increases are statistically significant, except on one occasion. Coupling ensemble classifiers with binary classification always yielded the highest accuracies. For our first data set, the highest accuracy was obtained with Adaboost and a 1-againstǃ strategy, 4.3% better than a single tree - for the second data set with the Random Forest approach and a 1-against-all strategy, 13.6% higher than a single tree.
|
|
