4 sonuçlar
Arama Sonuçları
Listeleniyor 1 - 4 / 4
Yayın Regularizing soft decision trees(Springer, 2013) Yıldız, Olcay Taner; Alpaydın, Ahmet İbrahim EthemRecently, 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 decision node which chooses one of the two. In this paper, we extend the original algorithm by introducing local dimension reduction via L-1 and L-2 regularization for feature selection and smoother fitting. We compare our novel approach with the standard decision tree algorithms over 27 classification data sets. We see that both regularized versions have similar generalization ability with less complexity in terms of number of nodes, where L-2 seems to work slightly better than L-1.Yayın Design and analysis of classifier learning experiments in bioinformatics: survey and case studies(IEEE Computer Soc, 2012-12) İrsoy, Ozan; Yıldız, Olcay Taner; Alpaydın, Ahmet İbrahim EthemIn many bioinformatics applications, it is important to assess and compare the performances of algorithms trained from data, to be able to draw conclusions unaffected by chance and are therefore significant. Both the design of such experiments and the analysis of the resulting data using statistical tests should be done carefully for the results to carry significance. In this paper, we first review the performance measures used in classification, the basics of experiment design and statistical tests. We then give the results of our survey over 1,500 papers published in the last two years in three bioinformatics journals (including this one). Although the basics of experiment design are well understood, such as resampling instead of using a single training set and the use of different performance metrics instead of error, only 21 percent of the papers use any statistical test for comparison. In the third part, we analyze four different scenarios which we encounter frequently in the bioinformatics literature, discussing the proper statistical methodology as well as showing an example case study for each. With the supplementary software, we hope that the guidelines we discuss will play an important role in future studies.Yayın Eigenclassifiers for combining correlated classifiers(Elsevier Science Inc, 2012-03-15) Ulaş, Aydın; Yıldız, Olcay Taner; Alpaydın, Ahmet İbrahim EthemIn practice, classifiers in an ensemble are not independent. This paper is the continuation of our previous work on ensemble subset selection [A. Ulas, M. Semerci, O.T. Yildiz, E. Alpaydin, Incremental construction of classifier and discriminant ensembles, Information Sciences, 179 (9) (2009) 1298-1318] and has two parts: first, we investigate the effect of four factors on correlation: (i) algorithms used for training, (ii) hyperparameters of the algorithms, (iii) resampled training sets, (iv) input feature subsets. Simulations using 14 classifiers on 38 data sets indicate that hyperparameters and overlapping training sets have higher effect on positive correlation than features and algorithms. Second, we propose postprocessing before fusing using principal component analysis (PCA) to form uncorrelated eigenclassifiers from a set of correlated experts. Combining the information from all classifiers may be better than subset selection where some base classifiers are pruned before combination, because using all allows redundancy.Yayın Statistical tests using hinge/ε-sensitive loss(Springer-Verlag, 2013) Yıldız, Olcay Taner; Alpaydın, Ahmet İbrahim EthemStatistical 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) however are trained to minimize the hinge (ε-sensitive) loss and hence they should not be assessed or compared in terms of the 0/1 (square loss) but with the loss measure they are trained to minimize. We discuss how the paired t test can use the hinge (ε-sensitive) loss and show in our experiments that doing that, we can detect differences that the test on error cannot detect, indicating higher power in distinguishing between the behavior of kernel-based classifiers (regressors). Such tests can be generalized to compare L > 2 algorithms.
