Yazar "Alpaydın, Ahmet İbrahim Ethem" için MF - Bildiri Koleksiyonu | Bilgisayar Mühendisliği Bölümü / Department of Computer Engineering listeleme
-
Budding trees
Yıldız, Olcay Taner; İrsoy, Ozan; Alpaydın, Ahmet İbrahim Ethem (IEEE Computer Soc, 2014-08-24)We propose a new decision tree model, named the budding tree, where a node can be both a leaf and an internal decision node. Each bud node starts as a leaf node, can then grow children, but then later on, if necessary, its ... -
Calculating the VC-dimension of decision trees
Yıldız, Olcay Taner; Aslan, Özlem; Alpaydın, Ahmet İbrahim Ethem (IEEE, 2009)We propose an exhaustive search algorithm that calculates the VC-dimension of univariate decision trees with binary features. The VC-dimension of the univariate decision tree with binary features depends on (i) the ... -
Multivariate statistical tests for comparing classification algorithms
Yıldız, Olcay Taner; Aslan, Özlem; Alpaydın, Ahmet İbrahim Ethem (Springer, Berlin, Heidelberg, 2011)The misclassification error which is usually used in tests to compare classification algorithms, does not make a distinction between the sources of error, namely, false positives and false negatives. Instead of summing ... -
Regularizing soft decision trees
Yıldız, Olcay Taner; Alpaydın, Ahmet İbrahim Ethem (Springer, 2013)Recently, we have proposed a new decision tree family called soft decision trees where a node chooses both its left and right children with different probabilities as given by a gating function, different from a hard ... -
Soft decision trees
İrsoy, Ozan; Yıldız, Olcay Taner; Alpaydın, Ahmet İbrahim Ethem (IEEE, 2012)We discuss a novel decision tree architecture with soft decisions at the internal nodes where we choose both children with probabilities given by a sigmoid gating function. Our algorithm is incremental where new nodes are ... -
Statistical tests using hinge/ε-sensitive loss
Yıldız, Olcay Taner; Alpaydın, Ahmet İbrahim Ethem (Springer-Verlag, 2013)Statistical tests used in the literature to compare algorithms use the misclassification error which is based on the 0/1 loss and square loss for regression. Kernel-based, support vector machine classifiers (regressors) ...