The primary objective of this research is toobtain an accurate FORECASTING model for the US presidentialelection. To identify a reliable model, artificial neuralnetworks (ANN) and support vector regression (SVR)models are compared based on some specified performancemeasures. Moreover, six independent variables such asGDP, unemployment rate, the president’ s approval rate, and others are considered in a stepwise regression toidentify significant variables. The president’ s approval rateis identified as the most significant variable, based onwhich eight other variables are identified and considered inthe model development. Preprocessing methods are appliedto prepare the data for the learning algorithms. The proposedprocedure significantly increases the accuracy of themodel by 50%. The learning algorithms (ANN and SVR)proved to be superior to linear regression based on eachmethod’ s calculated performance measures. The SVRmodel is identified as the most accurate model among theother models as this model successfully predicted theoutcome of the election in the last three elections (2004, 2008, and 2012). The proposed approach significantlyincreases the accuracy of the forecast.

Industrial FORECASTING is a top-echelon researchdomain, which has over the past several years experiencedhighly provocative research discussions. The scope of thisresearch domain continues to expand due to the continuousknowledge ignition motivated by scholars in the area. So, more intelligent and intellectual contributions on currentresearch issues in the accident domain will potentially sparkmore lively academic, value-added discussions that will be ofpractical significance to members of the safety community. Inthis communication, a new grey– fuzzy– Markov time seriesmodel, developed from nondifferential grey interval analyticalframework has been presented for the first time. Thisinstrument forecasts future accident occurrences under timeinvarianceassumption. The actual contribution made in thearticle is to recognise accident occurrence patterns anddecompose theminto grey state principal pattern components. The architectural framework of the developed grey– fuzzy– Markov pattern recognition (GFMAPR) model has fourstages: fuzzification, smoothening, defuzzification andwhitenisation. The results of application of the developednovel model signify that FORECASTING could be effectivelycarried out under uncertain conditions and hence, positions themodel as a distinctly superior tool for accident FORECASTINGinvestigations. The novelty of thework lies in the capability ofthe model inmaking highly accurate predictions and forecastsbased on the availability of small or incomplete accident data.

The concept of super line graph was introduced in the year 1995 by Bagga, Beineke and Varma. Given a graph with at least r edges, the super line graph of index r, Lr(G), has as its vertices the sets of r-edges of G, with two adjacent if there is an edge in one set adjacent to an edge in the other set. The line completion number lc(G) of a graph G is the least positive integer r for which Lr(G) is a complete graph. In this paper, we  nd the line completion number of grid graph Pn  Pm for various cases of n and m.

In this paper some properties of a generalization of Fibonacci sequence are investigated. Then we solve the Diophantine equations x2±kxy-y2±x=0, where k is a positive integer, and describe the structure of solutions.

Many studies have been conducted in the last decades on cellular polypropylene (Cell-PP) films. Most of them focus on theoptimisation of the material for sensor applications. Processed under Gas Diffusion Expansion (GDE), Cell-PP films showhigh piezoelectric activity and low stiffness/density, properties that make them ideal for sensors. GDE increases the heightand decreases the length over height ratio (aspect ratio) of individual voids within the material. This change in voidmorphology, and eventually stiffness, results in a nonlinear piezoelectric response of these materials. In this study, a Cell-PP sample was tested under static, quasi-static and low-frequency compressive stress. The main aim is to evaluate itsmechanical and piezoelectric properties in the nonlinear region of its response over strain. The load– deflection curves aswell as the piezoelectric responses were obtained for stresses up to 270 kPa (engineering strain close to 0. 26). It is shownthat both the magnitude of the initial load and the strain rate have a critical effect on the creep/stress relaxation of the filmand eventually on its piezoelectric response. Finally, it is shown that under dynamic conditions, and for the sameengineering strain region, it is more relevant to present the piezoelectric response, in terms of strain rather than stress.

Facility location problem (FLP) is a mathematicalway to optimally locate facilities within a set of candidatesto satisfy the requirements of a given set of clients. This study addressed the uncapacitated FLP as it assuresthat the capacity of every selected facility is finite. Thus, even if the demand is not known, which often is the case, inreality, organizations may still be able to take strategicdecisions such as locating the facilities. There are differentapproaches relevant to the uncapacitated FLP. Here, thecuckoo search via Le´ vy flight (CS-LF) was used to solvethe problem. Though hybrid methods produce betterresults, this study employed CS-LF to determine first itspotential in finding solutions for the problem, particularlywhen applied to a real-world problem. The method wasapplied to the data set obtained from a department store inDavao City, Philippines. Results showed that applying CSLFyielded better facility locations compared to particleswarm optimization and other existing algorithms. Although these results showed that CS-LF is a promisingmethod to solve this particular problem, further studies onother FLP are recommended to establish a strong foundationof the capability of CS-LF in solving FLP.

In the present study, the effect of size distribution of magnetite nanoparticles in a PVDF matrix on the magnetic propertiesof PVDF? Fe3O4 nanocomposites was experimentally and theoretically investigated. The size distribution of nanoparticlesin polymer matrix and morphology of the nanocomposites were studied by the means of scanning electron microscopyand atomic force microscopy. It was found that when the Fe3O4 nanoparticles are introduced into the polymermatrix, their coagulation takes place. The increase in the size of the particles depends on their concentration in the polymermatrix, the type of polymer (polar, non-polar, its viscosity, etc. ), reaction temperatures, etc. In addition, when Fe3O4nanoparticles are introduced into the polymer network, the oxidation of the surface layer of particles occurs and themagnetic size decreases. Consequently, the reduced magnetic properties may also be observed. The hysteresis loops havebeen recorded in small magnetic field range. It was found that the magnetic hysteresis parameters depend on the size andconcentration of Fe3O4 nanoparticles. Theoretical calculations were compared with experimental results obtained fromM(H) measurements. The reasons of differences between theoretical and experimental results have been explained.

A cooperativity is observed between CH/p and hydrogen bond interactions in HF···Py^X1benX2 complexes. The cooperativity is more influenced by the inductive effects in comparison with the resonance effects. Increase in the electron withdrawing nature of substituents is accompanied by decrease in the cooperativity energy (Ecoop). The vital effect of the reactivity centre on the Ecoop is determined through the good relationship between spara+ (the substituted constant estimated for the para placed reactivity center) and Ecoop values.