Competitive Randomized Nonlinear Prediction Under Additive Noise

We consider sequential nonlinear prediction of a bounded, real-valued and deterministic signal from its noise-corrupted past samples in a competitive algorithm framework. We introduce a randomized algorithm based on context-trees . The introduced algorithm asymptotically achieves the performance of the best piecewise affine model that can both select the best partition of the past observations space (from a doubly exponential number of possible partitions) and the affine model parameters based on the desired clean signal in hindsight. Although the performance measure including the loss function is defined with respect to the noise-free clean signal, the clean signal, its past samples or prediction errors are not available for training or constructing predictions. We demonstrate the performance of the introduced algorithm when applied to certain chaotic signals.      

Determination of bioclimatic comfort in three different land uses in the city of Erzurum,Turkey

This study was carried out to determine the human bioclimatic conditions in rural, urban and urban forest areas in the conditions of the city of Erzurum, where an extreme continental climate type prevails. Data were obtained over a 10-month period and human bioclimatic conditions in these three different land-use types were evaluated using human bioclimatic indices, suitable for the data obtained. In the study, thermohygrometric index (THI), which assesses air temperature and relative humidity, and “beer garden days” index, which uses the days when temperature at 21:00 is over 20 °C, were used. While in these three areas “hot” and “comfort” ranges were determined to be 10% of the period, the number of “beer garden” days was only 20 days in rural, 15 days in urban forest and 18 days in urban areas of Erzurum. Consequently, it has been determined that the most suitable area for the human comfort in the conditions of Erzurum is in the urban area which is followed by the urban forest and the rural areas, both for each observation time and across the whole period.     

Uncertainty in risk of highway bridges assessed for integrated seismic and flood hazards

Probabilistic risk assessment for bridges under natural hazards is of great interest to engineers for the development of risk mitigation strategies and implementation plans. The present study evaluates risk of an existing highway bridge in California, USA, under the integrated impact of regional seismic and flood hazards. A sensitivity study combining tornado diagram and first-order second moment reliability analyses is conducted to screen significant uncertain parameters to which bridge response is mostly sensitive. A rigorous uncertainty analysis, employing random sampling and Monte Carlo simulation techniques, is performed to obtain variations in fragility and risk curves of the bridge. Observed variations in risk curves at various risk levels are quantified through 90% confidence intervals and coefficients of variation (COV) of risk. It is observed that uncertainty in the estimated risk increases due to the presence of flood hazard at the bridge site, although mean risk does not vary with flood hazard level. Research outcome signifies that the variation in risk due to parameter uncertainty and varied flood hazard level should not be ignored to ensure bridge safety under the stated multi-hazard condition.     

Super porous α-, β-, γ-cyclodextrin cryogels with high active agent loading and controllable release profiles

Cyclodextrins (CDs) are truncated cone-like structures that are natural cyclic oligosaccharides. Here, a simple preparation method for super porous poly(α-CD), poly(β-CD), and poly(γ-CD) cryogels crosslinking with divinyl sulfone at 150%, 100% and 125% mole ratios with respect to the α-CD, β-CD, and γ-CD molecules, respectively, under cryogenic conditions, is reported. The interconnected homogeneous pore distribution of CD-based cryogels with pore sizes in the range of 5–100 μm is confirmed by SEM analysis. The CD-based cryogels weighing 10 mg are determined as hemocompatible with <1% hemolysis ratios and >79% blood clotting indexes; whereas the same materials weighing 1 mg are biocompatible with >75% cell viability on L929 fibroblasts. Additionally, active agent adsorption/delivery efficiencies of CD-based cryogels utilizing two active agents, Bisphenol A (BPA, a carcinogenic compound) and Curcumin (CUR, a polyphenolic compound), are individually evaluated. It was revealed that p(γ-CD) cryogels exhibited the highest active agent loading capacity for BPA, 87 ± 13 mg/g, whereas p(α-CD) cryogels showed the highest loading capacity for CUR, 136 ± 4 mg/g. Moreover, the active agent release from p(α-CD), p(β-CD), and p(γ-CD) cryogel networks at pH 7.4 and 37°C were determined as 40.6 ± 2, 35.3 ± 2, and 34 ± 1 mg/g for BPA, and 1.07 ± 0.2, 1.27 ± 0.1, and 1.37 ± 0.1 mg/g for CUR, respectively, within 96 h.     

The effect of recycling number on the mechanical,chemical, thermal,and rheological properties of PBT/PC/ABS ternary blends: With and without glass‐fiber

The recycling possibilities of poly(butylene terephthalate)/polycarbonate/acrylonitrile–butadiene–styrene (PBT/PC/ABS) ternary blend with and without glass‐fiber content were investigated using repeated injection molding process. In this study, PBT/PC/ABS ternary blends were reprocessed at five times and the results were presented after each recycling process. The recycling possibility of PBT/PC/ABS ternary blend was evaluated by measuring the mechanical, chemical, thermal, and rheological properties. Mechanical properties were determined by the tensile strength, yield strength, strain at break, elastic modulus, impact strength, flexural strength, and flexural modulus. Chemical and thermal properties were evaluated by Fourier transform infrared spectroscopy, differential scanning calorimetry, thermal gravimetric analysis, and scanning electron microscopy. Rheological properties of the ternary blends were studied by melt flow index measurement. From the results, it was found that mechanical properties of recycled composites were better than virgin PBT/PC/ABS ternary blends. POLYM. COMPOS., 35:2074–2084, 2014. © 2014 Society of Plastics Engineers     

Evaluation of surface settlements in the Istanbul metro in terms of analytical, numerical and direct measurements

In urban areas, the use of pre-support in shallow tunnels is increasing in order to improve the excavation stability and reduce ground settlements induced by the tunneling. This paper describes the effects of pipe roofing on surface settlements and evaluates the settlements associated with the twin tunnels in the Istanbul metro in terms of numerical, semi-empirical and measured values. FLAC3D was used for the numerical modeling and the method suggested by Herzog for the semi-empirical solution. The numerical results indicate that the tunnel roof formed by the pipe roofing provides a restraining effect, reducing deformation and ground surface settlement by up to 65?%. Herzog’s model yields higher maximum surface settlements than those observed. When modified to include the effect of pipe roofing and the distance between the pipes however, there is good agreement between the direct measurements and the values predicted by the numerical and modified Herzog models.     

Planar shape representation and matching under projective transformation

This paper introduces a new representation for planar objects which is invariant to projective transformation. Proposed representation relies on a new shape basis which we refer to as the conic basis. The conic basis takes conic-section coefficients as its dimensions and represents the object as a convex combination of conic-sections. Pairs of conic-sections in this new basis and their projective invariants provides the proposed view invariant representation. We hypothesize that two projectively transformed versions of an object result in the same representation. We show that our hypothesis provides promising recognition performance when we use the nearest neighbor rule to match projectively deformed objects.     

Predictors of attitudinal and behavioral outcomes in mobile advertising: A field experiment

The study examines consumer responses to SMS-based mobile advertising campaigns using a field experiment with actual behavioral responses. Specifically, the effects of message characteristics (i.e., incentives and prior permission) and individual differences of message recipients (i.e., content involvement, prior experience with the mobile medium, and medium-fit perceptions) on the attitudinal and behavioral outcomes of such campaigns are investigated. Perceived intrusiveness of the message and attitude toward the campaign are conceptualized as key attitudinal reactions triggered by a mobile advertising message, which in turn influence the behavioral responses generated by the campaign. Findings suggest that, in comparison to permission and incentive, individual differences are stronger determinants of responses to mobile advertising campaigns. Additionally, intriguing insights with respect to the effectiveness of different types of incentives are provided.