In traditional classification problems, the reference needed for training a classifier is given and considered to be absolutely correct. However, this does not apply to all tasks. In emotion recognition in non-acted speech, for instance, one often does not know which emotion was really intended by the speaker. Hence, the data is annotated by a group of human labelers who do not agree on one common class in most cases. Often, similar classes are confused systematically. We propose a new entropy-based method to evaluate classification results taking into account these systematic confusions. We can show that a classifier which achieves a recognition rate of "only" about 60 % on a four-class-problem performs as well as our five human labelers on average.

No abstract

A pulsed electron spin resonance (ESR) microimaging system operating at the Q-band frequency range is presented. The system includes a pulsed ESR spectrometer, gradient drivers, and a unique high-sensitivity imaging probe. The pulsed gradient drivers are capable of producing peak currents ranging from ∼9 A for short 150 ns pulses up to more than 94 A for long 1400 ns gradient pulses. Under optimal conditions, the imaging probe provides spin sensitivity of ∼1.6 × 10(8) spins∕√Hz or ∼2.7 × 10(6) spins for 1 h of acquisition. This combination of high gradients and high spin sensitivity enables the acquisition of ESR images with a resolution down to ∼440 nm for a high spin concentration solid sample (∼10(8) spins∕μm(3)) and ∼6.7 μm for a low spin concentration liquid sample (∼6 × 10(5) spins/μm(3)). Potential applications of this system range from the imaging of point defects in crystals and semiconductors to measurements of oxygen concentration in biological samples.

Abstract-We have developed components of an automated system that understands and follows navigational instructions. The system has prior knowledge of the geometry and landmarks of specific maps. This knowledge is exploited to infer complex paths through maps based on natural language descriptions. The approach is based on an analysis of verbal commands in terms of elementary semantic units that are composed to generate a probability distribution over possible spatial paths in a map. An integration mechanism based on dynamic programming guides this language-to-path translation process, ensuring that resulting paths satisfy continuity and smoothness criteria. In the current implementation, parsing of text into semantic units is performed manually. Composition and interpretation of semantic units into spatial paths is performed automatically. In the evaluations, we show that the system accurately predicts the speakers' intended meanings for a range of instructions. This paper provides building blocks for a complete system that, when combined with robust parsing technologies, could lead to a fully automatic spatial language interpretation system. Index Terms-Human-machine interaction, natural language processing, navigational instructions, spatial language understanding.

Interfacial reactions, phase formation, microstructure, and composition, as functions of heat treatments (400–800 °C) were investigated in Ni90Ti10 alloy thin film coevaporated on an n-type 6H-SiC (0001) single-crystal substrate. The study was carried out with the aid of Auger electron spectroscopy, x-ray diffraction, and analytical transmission electron microscopy. The interaction was found to begin at 450 °C. Ni and C are the dominant diffusing species. The reaction zone is divided into three layers. In the first layer, adjacent to the SiC substrate, the presence of Ni-rich silicide, Ni2Si, and C precipitates, was observed. The second layer is composed mainly of TiC, while the third consists of Ni2Si. This composite structure, consisting of the silicide as a low resistivity ohmic contact, and of the carbide as a diffusion barrier, promises high-temperature stability crucial to ohmic contact development for SiC technology. Factors controlling phase formation in the Ni–Ti/SiC system are discussed.