The performance of m-out-of-n bagging with and without replacement in terms of the sampling ratio (m/n) is analyzed. Standard bagging uses resampling with replacement to generate bootstrap samples of equal size as the original training set mwor = n. Without-replacement methods typically use half samples mwr = n/2. These choices of sampling sizes are arbitrary and need not be optimal in terms of...

We propose a learning framework that actively explores creation of face space(s) by selecting images that are complementary to the images already represented in the face space. We also construct ensembles of classifiers learned from such actively sampled image sets, which further provides improvement in the recognition rates. We not only significantly reduce the number of images required in the...

The demand for humanoid robots as service robots for everyday life has increased during the last years. The processing power of the hardware and the development of complex software applications allows the realization of “natural” human-robot interaction. One of the important topics of natural interaction is the detection of emotions which enables the robot to react appropriate to the emotional ...

One of the most important directions in improvement of the datamining and knowledge discovery methods is the integration of the multiple classification techniques based on ensembles of classifiers. An integration technique should solve the problem of estimation and selection of the most appropriate component classifiers for an ensemble. We discuss an advanced dynamic integration of multiple cla...

We studied classification of human ECGs labelled as normal sinus rhythm, ventricular fibrillation and ventricular tachycardia by means of support vector machines in different representation spaces, using different observation lengths. ECG waveform segments of duration 0.5-4 s, their Fourier magnitude spectra, and lower dimensional projections of Fourier magnitude spectra were used for classific...

This paper presents a novel tree classifier for complex object detection tasks together with a general framework for real-time object tracking in videos using the novel tree classifier. A boosted training algorithm with a clustering-and-splitting step is employed to construct branches in the nodes recursively, if and only if it improves the discriminative power compared to a single monolithic n...

The scenario of classification with imbalanced datasets has gained a notorious significance in the last years. This is due to the fact that a large number of problems where classes are highly skewed may be found, affecting the global performance of the system. A great number of approaches have been developed to address this problem. These techniques have been traditionally proposed under three ...

a new parametric model was developed for predicting cut point of hydraulic classifiers. the model directly uses operating parameters including pulp flowrate, feed particle size characteristics, pulp solids content, solid density and particles retention time in the classification chamber and also covers uncontrollable errors using calibration constants. the model applicability was first verified...