Nonlinear transformation of term frequencies for term weighting in text categorization

In automatic text categorization, the influence of features on the decision is set by the term weights which are conventionally computed as the product of term frequency and collection frequency factors. The raw form of term frequencies or their logarithmic forms are generally used as the term frequency factor whereas the leading collection frequency factors take into account the document frequency of each term. In this study, it is firstly shown that the best-fitting form of the term frequency factor depends on the distribution of term frequency values in the dataset under concern. Taking this observation into account, a novel collection frequency factor is proposed which considers term frequencies. Five datasets are firstly tested to show that the distribution of term frequency values is task dependent. The proposed method is then proven to provide better F₁ scores compared to two recent approaches on majority of the datasets considered. It is confirmed that the use of term frequencies in the collection frequency factor is beneficial on tasks which does not involve highly repeated terms. It is also shown that the best F₁ scores are achieved on majority of the datasets when smaller number of features are considered.     

Cooperative communications with multilevel/AES-SD4-CPFSK in wireless sensor networks

In this paper, a new joint multilevel data encryption and channel coding mechanism is proposed, which is called ??multilevel/advanced encryption standard?Csystematic distance 4?Ccontinuous phase frequency shift keying?? (ML/AES-SD4-CPFSK). In the proposed scheme, we have not only taken advantage of spatial diversity gains but also optimally allocated energy and bandwidth resources among sensor nodes as well as providing high level of security and error protection for cooperative communications in wireless sensor networks. Relay protocols of cooperative communications, such as amplify-and-forward and decode-and-forward with/without adversary nodes, have been studied for 4CPFSK, 8CPFSK, and 16CPFSK of ML/AES-SD4-CPFSK. We have evaluated the error performances of multilevel AES for data encryption, multilevel SD-4 for channel coding, and various CPFSK types for modulation utilizing cooperative communications in wireless sensor networks. According to computer simulation results, significant diversity gain and coding gain have been achieved. As an example, bit error rate (BER) performance of 10^?5 value has been obtained at a signal-to-noise ratio (SNR) of ?6?dB for SD-4-CPFSK scheme in a compared related journal paper, whereas in our proposed system, we have reached the same BER value at a SNR of ?23?dB with amplify-and-forward with direct path signal protocol in 16-level AES, two-level SD-4 coded 16CPFSK, and at the same time, we have reached the same BER value at a SNR of ?22?dB with amplify-and-forward without direct path signal protocol in 16-level AES, two-level SD-4 coded 16CPFSK.     

Investigation of the effects of geometrical parameters on heat transfer from buried finned pipes

Heat transfer from finned pipes buried in the soil has been investigated using experimental, analytical and numerical techniques. The experimental setup consists of a ground source heat pump system and two separate ground heat exchangers. Constant surface temperature approach based on Carslaw and Jaeger’s theory study was used for calculating the heat transfer from the pipes analytically. The problem was modeled and solved using a CFD program numerically. Effects of the fin number, fin length and simulation time on the heat transfer rate have been investigated and compared with bare tube. A modified analytical formula was also proposed for finned pipes in this study.     

Impact of non-normal random effects on inference by multiple imputation: A simulation assessment

Multivariate extensions of well-known linear mixed-effects models have been increasingly utilized in inference by multiple imputation in the analysis of multilevel incomplete data. The normality assumption for the underlying error terms and random effects plays a crucial role in simulating the posterior predictive distribution from which the multiple imputations are drawn. The plausibility of this normality assumption on the subject-specific random effects is assessed. Specifically, the performance of multiple imputation created under a multivariate linear mixed-effects model is investigated on a diverse set of incomplete data sets simulated under varying distributional characteristics. Under moderate amounts of missing data, the simulation study confirms that the underlying model leads to a well-calibrated procedure with negligible biases and actual coverage rates close to nominal rates in estimates of the regression coefficients. Estimation quality of the random-effect variance and association measures, however, are negatively affected from both the misspecification of the random-effect distribution and number of incompletely-observed variables. Some of the adverse impacts include lower coverage rates and increased biases.     

TN approximation on the critical size of time-dependent,one-speed and one-dimensional neutron transport problem with anisotropic scattering

The criticality problem is studied based on one-speed time-dependent neutron transport theory, for a uniform and finite slab, using the Marshak boundary condition. The time-dependent neutron transport equation is reduced to a stationary equation. The variation of the critical thickness of the time-dependent system is investigated by using the linear anisotropic scattering kernel together with the combination of forward and backward scattering. Numerical calculations for various combinations of the scattering parameters and selected values of the time decay constant and the reflection coefficient are performed by using the Chebyshev polynomials approximation method. The results are compared with those previously obtained by other methods which are available in the literature.     

Multilayer Graphene Broadband Terahertz Modulators with Flexible Substrate

Fabrication of terahertz modulators as simple devices with high modulation depth across a broad bandwidth is still very challenging. In this study, four different chemical vapor deposition grown multilayer graphene (MLG) modulators based on MLG/ionic liquid/gold sandwich structures have been investigated. Flexible substrates (PVC and PE) were chosen as host materials, and devices were fabricated with three different thicknesses. The resultant MLG devices can be operated at low voltages between 0 and 3.4 V providing nearly complete modulation between 0.2 and 1.5 THz with low insertion losses. Even with such low gate voltages, the devices have been doped significantly inducing 7–11-fold improvement in their sheet conductivities depending on device thickness. In addition, sheet conductivity has been improved more than three times as the graphene layer number increased from 30 to 100. With the demonstration of promising device performances, the proposed modulators can be potential candidates for applications in terahertz and related optoelectronic technologies.     

Heat pipes thermoelectric solar collectors for energy applications

This study investigates the load characteristics of heat pipe thermoelectric solar collector (HPTSC) in practice. Heat pipe thermoelectric solar collector converts the heat generated by the Sun directly into electrical energy and produces hot water as well. The maximum power in HPTSC is obtained when the internal resistance of the thermoelectric module is equal to the load resistance. It has been observed to be possible to produce both hot water and electricity by improving available solar collectors or producing new generation HPTSC. While it is possible to generate an electrical power of 160 W from a HPTSC of one square meter using the thermoelectric method, the power produced with an average photovoltaic panel with the same area is only 132 W. Accordingly, HPTSC is a superior alternative not only to available solar collectors, but also to available PV panels. HPTSC, involving three different technologies, is environmentally friendly and certainly a product that allows for more efficient use of solar energy.     

Diversity and Software Development

Scott E. Page's diversity framework provides useful insights regarding software engineering research, practice, and education. This editorial discusses the concepts and implications of diversity in this context.     

Effect of reuterin produced by Lactobacillus reuteri on the surface of sausages to inhibit the growth of Listeria monocytogenes and Salmonella spp

Reuterin is a bacteriocin produced by some strains of Lactobacillus reuteri. The strain used in this study was isolated from raw milk from a dairy farm nearby Ankara. Beef sausage is a long years produced bratwurst style meat product in Turkey, as well as in some other countries in the Mediterranean region. Sausages are produced by raw meat; sometimes lactic starter cultures are added or spontaneous fermentation is employed. The production and storage conditions of the product promotes the growth of Listeria monocytogenes and Salmonella spp. Although nitrate is added as an antimicrobial substance against many pathogens, sometimes however nitrate application is not preventive enough on the surface because of the natural film around the sausages. Since most of the contaminations take place at post production steps, pathogenic growth is more effective on the surface of the sausages in refrigerated conditions. In this study, reuterin was applied to the surface of the sausages in order to prevent the growth of these two pathogens along with nitrate used as an additive in the product. Reuterin has inhibited the growth of L. monocytogenes considerably but not of Salmonella spp. on the surface of the sausages.     

Packet Reservation Window Multiple Access for Microcellular Voice/Data Transmission

We propose a new packet reservation multiple access (PRMA) scheme for the joint transmission of voice and data traffics in a microcellular medium. The collision resolution protocol within the system is based on a modification of the window random access algorithm, which has superior properties compared to the conventional slotted Aloha. The proposed algorithm, which we call packet reservation window multiple access (PRWMA), works in distinct modes for voice and data without prioritization, and the user performs slightly different operations depending on the information type. Simulation results show that PRWMA outperforms PRMA by a significant margin in terms of voice user capacity.