Hubness-aware shared neighbor distances for high-dimensional k-nearest neighbor classification

Learning from high-dimensional data is usually quite challenging, as captured by the well-known phrase curse of dimensionality. Data analysis often involves measuring the similarity between different examples. This sometimes becomes a problem, as many widely used metrics tend to concentrate in high-dimensional feature spaces. The reduced contrast makes it more difficult to distinguish between close and distant points, which renders many traditional distance-based learning methods ineffective. Secondary distances based on shared neighbor similarities have recently been proposed as one possible solution to this problem. However, these initial metrics failed to take hubness into account. Hubness is a recently described aspect of the dimensionality curse, and it affects all sorts of $k$ -nearest neighbor learning methods in severely negative ways. This paper is the first to discuss the impact of hubs on forming the shared neighbor similarity scores. We propose a novel, hubness-aware secondary similarity measure $simhub_s$ and an extensive experimental evaluation shows it to be much more appropriate for high-dimensional data classification than the standard $simcos_s$ measure. The proposed similarity changes the underlying $k$ NN graph in such a way that it reduces the overall frequency of label mismatches in $k$ -neighbor sets and increases the purity of occurrence profiles, which improves classifier performance. It is a hybrid measure, which takes into account both the supervised and the unsupervised hubness information. The analysis shows that both components are useful in their own ways and that the measure is therefore properly defined. This new similarity does not increase the overall computational cost, and the improvement is essentially ‘free’.     

A personal perspective on photowork: implicit human–computer interaction for photo collection management

In the age of digital photography, the amount of photos we have in our personal collections has increased substantially along with the effort needed to manage these new, larger collections. This issue has already been addressed in various ways: from organization by meta-data analysis to image recognition and social network analysis. We introduce a new, more personal perspective on photowork that aims at understanding the user and his/her subjective relationship to the photos. It does so by means of implicit human–computer interaction, that is, by observing the user’s interaction with the photos. In order to study this interaction, we designed an experiment to see how people behave when manipulating photos on a tablet and how this implicitly conveyed information can be used to aid photo collection management.     

Disjunction property and complexity of substructural logics

We systematically identify a large class of substructural logics that satisfy the disjunction property (DP), and show that every consistent substructural logic with the DP is PSPACE-hard. Our results are obtained by using algebraic techniques. PSPACE-completeness for many of these logics is furthermore established by proof theoretic arguments.     

Heat distribution in thermally modified timber

The heat distribution during thermal modification of beech (Fagus sylvatica L.) and spruce (Picea abies L. Karst) wood with dimensions of 80 × 80 × 200 mm³ has been investigated. Heat distribution was continuously measured by thermocouples in longitudinal and transverse directions. Significant temperature gradients occur in the initial phase of the process as well as during the modification phase (set temperature 200 °C for 3 h), where ongoing chemical reactions in wood were taking place. Thus, the temperature in the investigated positions increased to 240 °C (beech) and 215 °C (spruce). The mentioned properties should be taken into account when optimizing the heat treatment process (quality control, energy savings, etc.).     

Qualitative and quantitative analysis of social network data intended for brand management

Wireless Networks - The purpose of the paper is to point out the importance of analysing data obtained from social media through the qualitative data analysis software. The main goal of conducted...     

Rubrene as an additive in M-phthalocyanine/fullerene organic solar cells

Photovoltaic devices made from metallo-phthalocyanine/fullerene (M-Pc/C₆₀) with 5,6,11,12-tetraphenylnaphthacene (rubrene) as an additive are fabricated and characterized. The effect of rubrene is examined for 4 different M-phthalocyanines – H₂Pc, ZnPc, ClInPc, and VOPc – to represent 4 different valencies of the central moiety of M-Pc. In each case, rubrene has shown a notable increase in the open circuit voltage and in the case of the ClInPc and VOPc results in an increase in the overall power conversion efficiency. Through measurement of external quantum efficiency, it is shown that the increased efficiency is due to increased V_oc and not due to the photocurrent contribution from the complementary absorption profile of rubrene. Finally, the photostability of rubrene-based cells is studied, showing that unencapsulated devices decay rapidly in air as a result of the formation of rubrene peroxide, as evidenced by significant decay of the UV–vis absorption and direct measurement of the cell performance over a time period as short as several minutes.     

Heteromorphic chloroindium phthalocyanine films for improved photovoltaic performance

A new structure incorporating multiple phases of chloroindium phthalocyanine (ClInPc) is fabricated and tested in photovoltaic devices. This so-called heteromorphic structure includes as-deposited and THF vapor treated ClInPc films to improve absorption and photovoltaic (PV) performance in devices. Absorption of the polymorphic phases of ClInPc are complementary and lead to improved current generation. Short circuit current is improved by over 70% using the heteromorphic structure, while power conversion efficiency (PCE) improves by more than 40% versus solely as-deposited devices. Advantages of the heteromorphic structure include broader spectral response, improved interfacial contact area and an intermediary open circuit voltage (V_oc).     

Propagation-invariant electromagnetic fields: Theory and experiment

Abstract

The vector concept of the propagation invariance is formulated to be applied to the stationary electromagnetic fields. The analysed exact solutions of the Maxwell equations are obtained on the basis of general propagation-invariant solutions of the scalar Helmholtz equation. A possible classification of the propagation-invariant fields based on quantities which appear in the complex Poynting theorem is proposed. The longitudinal periodicity of the electromagnetic field obtained due to the superposition of two Bessel beams is verified by a simple experiment.     

Silver‐Catalyzed Cyclization of Propargylic Amides to Oxazolines

A ligand‐accelerated effect is observed in the cyclization of propargylic amides catalyzed by bis(pyridyl)silver(I) complexes, with an unexpected reversal of electronic demand to the analogous N H addition reaction. The catalyst was found to be effective for internal alkyne substrates, offering exclusive selectivity for the 5‐exo‐dig product. Differences in selectivity profile between gold‐ and silver‐catalyzed processes are highlighted and discussed.