The Effect of Target Vector Selection on the Invariance of Classifier Performance Measures

In this paper, the multiclass supervised training problem is considered when a discrete set of classes is assumed. Upon generating affine models for finite data sets, we have observed the invariance of certain measures of performance after a trained classifier has been presented with test data of unknown classification. Specifically, after constructing mappings between training vectors and their desired targets, the class membership and ranking of test data has been found to remain either invariant or close to invariant under a transformation of the set of target vectors. Therefore, we derive conditions explaining how this type of invariance can arise when the multiclass problem is phrased in the context of linear networks. A bioinformatics example is then presented in order to demonstrate various principles outlined in this work.     

Detection of low levels of trichloroethylene vapor with Raman spectrometry

The response of a novel fiber-optic Raman probe to low levels of trichloroethylene (TCE) vapors is characterized. The detection limit of the current probe for TCE vapor is 34 mg/L, and the probe exhibits a fully reversible response. The probe uses an organic-polymer, low-density polyethylene to concentrate TCE vapors in the optical path of the fiber-optic Raman spectrometer. The relative standard deviation for measurement of 677 mg/L of TCE in the vapor is 0.3%.     

Characterization of laser hardened steels by laser induced ultrasonic surface waves

Ultrasonic surface waves are suitable for the characterization of surface hardened materials. This is shown on laser hardened turbine blades. The martensitic microstructure within the surface layer of surface hardened steels has a lower surface wave propagation velocity than the annealed or normalized substrate material. Because the propagation velocity depends on the ratio of layer thickness to wavelengthd/, its measurement allows the determination of the hardening depth. If the surface wave frequency is high enough, the surface wave propagates mainly within the hardened layer. A correlation of the surface wave velocity to the surface hardness has been found. Because the variation of the surface velocity in hardened steels is small, a high measurement accuracy is necessary to obtain the interesting hardening parameters with sufficient certainty. Therefore, a measuring arrangement has been developed where laser pulses, guided by optical fibers to the surface hardened structure, generate simultaneously surface wave pulses at two different positions. The two ultrasonic pulses are received by a piezoelectric transducer. The surface wave velocity is obtained from the time delay between these pulses which is determined by the cross-correlation method. To evaluate simultaneously surface waves with different penetration depths from the same signal acquisition, digital filtering has been used in connection with the cross-correlation.     

Protecting and preserving clear barrier layers for flexible packaging materials In‐line thermal evaporation of organic topcoat layer

The end market for transparent flexible barrier films is larger than for metallized films. Presently, the market is still dominated by polymeric barrier layers but the used chemicals may be harmful for the environment. An alternative would be transparent thin layers deposited by vacuum deposition techniques using reactive processes. Ceramic materials like silicon oxide or aluminum oxide are used having a film thickness of just ～10 nm, a coating uniformity of +/?5% across and along the film at a barrier performance below 2.0 sccm/m²d for oxygen transmission rate (OTR) and below 1.0 g/m²d for water vapor transmission rate (WVTR) on PET substrates. In this paper, details will be provided about the deposition processes for these barrier layers using thermal evaporation, plasma‐assisted thermal evaporation as well as deposition by electron beam evaporation. An important factor for these high barrier transparent coatings is also to withstand the downstream processes in the whole packaging stream like slitting, lamination, printing etc. One solution is to protect the barrier layers by a Topcoat. For example, off‐line deposition of lacquers is used in field but the market penetration is low due to high process and material costs. An in‐situ Topcoat deposition is a smart solution to overcome this issue saving time and costs. Such an approach will be also described in the presentation and the impact on the performance of the final package will be discussed.     

New aspects of transversal1H-NMR relaxation in natural rubber vulcanizates

Summary Structure and dynamics of unfilled and carbon black filled natural rubber vulcanizates were investigated by means of transversal¹H-NMR-relaxation. A simple theoretical model including different motions in polymer network was extended by the discussion of an anisotropy parameter of segmental motion not only for the intercrosslink chains but also for the dangling ends. It makes characterization of the filling state of the network possible to approximate. The effect of increasing correlation times _f of the fast anisotropic local motion with reduction of the molecular mass of intercrosslink chains which was observed, but not cleared up in previous investigation is evidently smaller by using this modified model in accordance with the expectation of a local motion. The determined values of M_c are in good agreement with the results of fits by using the unmodified relaxation function.¹³C-NMR-spectra and stress-strain-experiments are leading in most cases to similar results, especially for M_c.     

Natural perspective projections for head-mounted displays

The display units integrated in today's head-mounted displays (HMDs) provide only a limited field of view (FOV) to the virtual world. In order to present an undistorted view to the virtual environment (VE), the perspective projection used to render the VE has to be adjusted to the limitations caused by the HMD characteristics. In particular, the geometric field of view (GFOV), which defines the virtual aperture angle used for rendering of the 3D scene, is set up according to the display field of view (DFOV). A discrepancy between these two fields of view distorts the geometry of the VE in a way that either minifies or magnifies the imagery displayed to the user. It has been shown that this distortion has the potential to affect a user's perception of the virtual space, sense of presence, and performance on visual search tasks. In this paper, we analyze the user's perception of a VE displayed in a HMD, which is rendered with different GFOVs. We introduce a psychophysical calibration method to determine the HMD's actual field of view, which may vary from the nominal values specified by the manufacturer. Furthermore, we conducted two experiments to identify perspective projections for HMDs, which are identified as natural by subjects--even if these perspectives deviate from the perspectives that are inherently defined by the DFOV. In the first experiment, subjects had to adjust the GFOV for a rendered virtual laboratory such that their perception of the virtual replica matched the perception of the real laboratory, which they saw before the virtual one. In the second experiment, we displayed the same virtual laboratory, but restricted the viewing condition in the real world to simulate the limited viewing condition in a HMD environment. We found that subjects evaluate a GFOV as natural when it is larger than the actual DFOV of the HMD--in some cases up to 50 percent--even when subjects viewed the real space with a limited field of view.     

Optimization problems in contracted tensor networks

We discuss the calculus of variations in tensor representations with a special focus on tensor networks and apply it to functionals of practical interest. The survey provides all necessary ingredients for applying minimization methods in a general setting. The important cases of target functionals which are linear and quadratic with respect to the tensor product are discussed, and combinations of these functionals are presented in detail. As an example, we consider the representation rank compression in tensor networks. For the numerical treatment, we use the nonlinear block Gauss?CSeidel method. We demonstrate the rate of convergence in numerical tests.     

Rechtliche Fragen zur Praxis IT-forensischer Analysen in Organisationen

Der Einsatz von Informationstechnologie (IT) in Organisationen bietet viele M?glichkeiten des Missbrauchs. Besch?ftigte k?nnen sich die besonderen Eigenschaften der IT zunutze machen, um sich etwa in gesch?ftssch?digender Weise Vorteile zu verschaffen. Gleichzeitig erm?glicht IT aber auch, verbotene Handlungen aufzuspüren und zu belegen (IT-Forensik). Der Beitrag beschreibt die technische Praxis von IT-forensischen Analysen und geht auf damit verbundene rechtliche Fragestellungen ein.