European Journal of Scientific Research

Volume 140 No 4
July, 2016

A Single Dose of Blueberry Leaf Extract Suppresses Serum Ethanol Levels after oral Ethanol Administration in Rats
352-365
Kaede Yamasaki, Hirotaka Nagatomo, Yuki Kawamura, Kazuhiro Sugamoto, Takanori Kai, Kozo Kamenaga, Masahiko Takeshita, Yukiharu Kikuchi, Yasushi Matsuura, Chizuko Yukizaki, Kazuo Nishiyama and Masao Yamasaki
Abstract:
The effect of blueberry leaf extract (BLEx) on alcohol metabolism was evaluated in Sprague-Dawley (SD) rats by determining the serum levels of ethanol and acetaldehyde, expression and activity of alcohol-related metabolism enzymes, and serum glutamic pyruvic transaminase (GPT) activity. A single dose ofBLEx significantly inhibited the elevation of serum ethanol and acetaldehyde after ethanol administration. Liver alcohol dehydrogenase (ADH) activity significantly increased in the ethanol control group and decreased in the BLEx group compared to the activities in the normal control and ethanol group, respectively. In addition, liver microsomal cytochrome P450 2E1 (CYP2E1) protein expression increased significantly in the ethanol control group but CYP2E1 expression was not affected by BLEx. The mRNA expression of alcohol-related metabolism enzymes in the liver did not differ significantly among the experimental groups. Ethanol administration did not increase serum GPT activity. Thus, BLEx could be useful forpreventing alcohol-related disorders by inhibiting ethanol absorption.
Keywords: blueberry leaf extract, alcohol metabolism, ethanol

Outlier-Eliminated Extreme Learning Machine for Classification Problems
366-376
Vimala Alexander and Pethalakshmi Annamalai
Abstract:
Extreme Learning Machine (ELM) is one of the faster learning approach used in single layer feed forward neural networks. The greater speed and the more accurate generalization output are the advantages of ELM. However, the presence of outliers might reduce the training performance of ELM. The literature shows that λ1-norm loss function is widely used for outlier detection, however, these methods are iterative and tend to adjust the model to learn along with the outliers. In this paper, a simple and robust method is presented to detect and eliminate the outliers from the data to improve training accuracy, hence the testing phase can also be enhanced. The experiments on benchmark datasets shows that the proposed approach not only enhances original ELM training, but also results in consistent and notable accuracy in handling data with outliers.
Keywords: Artificial Neural Networks, Backpropagation, Extreme Learning Machine, Outliers.

Pétrographie et Géochimie du Massif Granitique de Parakou (NE – Bénin)
377-393
KWEKAM Maurice, TAIROU Mahaman Sani, CHALA Dominique, WENMENGA Urbain, FLEURY Jules, AFFATON Pascal and MORIBA Djibril Aliou
Abstract:
The Parakou Massif, in the NE-Benin, represents one of the first granitic intrusions in the Benino-Nigerian Metacraton frontal part. It is mainly made up of porphyritic biotite bearing leucogranites. These rocks are more or less structured by a Pan-African sinistral shearing which is superposed on a magmatic fluidal structure.
Petro-geochemical studies of samples from the Parakou Granitic Massif shows that it is composed of rocks belong to the calc-alkaline series. These rocks are I type syn-cinematic granitoids, slightly peraluminous, resulting from magmatic differentiation by fractional crystallization. They are mainly monzogranites and probably correspond to mixed origin magma with the two crustal and mantle derived components. All the petro-structural and geochemical characteristics of the Parakou Granitic Massif reflect post-collisional geodynamic context setting.
Keywords: Pan-African, Dahomeyide, NE-Benin, syn-cinematic granite, post-collision

Enhanced Clustering Based Approach for Predicting Next Web Page using Web Log Data- A Study
394-401
Jothish Chembath and E.J.Thomson Fredrik
Abstract:
Increase in number of web pages has necessitated next web page prediction a mandatory step during web usage mining in order to increase user satisfaction, retention and profitability. A next web page prediction system consists of two major steps, namely, preprocessing and prediction. In this work, a clustering method is proposed to cluster pages in a session with respect to number of visits. The inclusion of cluster algorithm can improve the performance of next web page prediction. For this purpose, the conventional K-Means algorithm is modified in three manners. The first is to automatically find K (number of clusters) value using ensemble method. The second is to automatically identify initial centroids. The third is to propose method that can reduce the number of distance calculations performed in order to reduce its complexity. This clustering algorithm is combined with longest common sequence algorithm to identify the future request of the user. Experimental results proved that the inclusion of clustering algorithm has improved the prediction performance and improves the prediction process when compared to longest common sequence algorithm and conventional K-Means algorithm.
Keywords: Next Page Prediction, Grouping Web Pages, K-Means Clustering, Longest Common Sequence Algorithm, Web Log Data, Web Usage Mining


Elemental Concentration of Natural Occurring Radioactive Materials in Soils Nearby Uranium Mine Sites of Erongo Region, Namibia
402-410
Zivuku Munyaradzi, Kgabi Nnnenesi Anna and Tshivhase Victor
Abstract:
The aim of this study was to determine the elemental concentrations of 238U, 232Th and 40K in soil samples and estimate the radiological hazards to the general public and the environment.A total of twenty soils samples were obtained from coastaltowns of Walvis Bay and Swakopmund (Erongo region), in the vicinity of uranium mines and these were analyzed for mass elemental concentration of radionuclides - uranium, thorium and potassium due to activities of U-238 and Th-232 and K-40 using HPGe gamma –ray spectrometer, followed by assessment of radiological hazards to the public and the environment.The mass elemental concentration in all soil samples collected from the two coastal towns varied from 1±0.31 to 16±3.00 mg kg-1for 238U, 4 ±1.14 to 32±4.74 mg kg-1for 232Th and 0.5±17 to 2±0.48 % for 40K. The mean values of radium equivalent index (Raeq), radiation hazard index (Hex) and total annual external gamma radiation dose were found to be 205 Bq kg-1, 0.56 and 0.11 mSvyr-1respectively.It was observed that there is a significant difference in the mass elemental concentration and distribution of NORMs between Walvis Bay and Swakopmund town. The estimated radiological impact assessment factors were lower than the 1 mSv yr-1 set by United Nations Scientific Committee on the Effect of Atomic Radiation (UNSCEAR) for the general public. The results obtained thus indicate there seems to be no potential radiological health hazard associated with the soils of Walvis Bay and Swakopmund towns. This shows that the risk due to radiation contamination in Erongo region is low.
Keywords: Radioactivity; soil contamination; health hazards

Evaluation des Performances de Neuf (09) Modèles Hydrologiques Pluie-débit Globaux sur le Bassin de la Mékrou à L’exutoire de Kompongou (Bénin)
411-424
Obada Ezéchiel, Alamou Adéchina Eric and Afouda Abel
Abstract:
Dans le but de contribuer à la mise en place d’un cadre adéquat de gestion de la ressource en eau dans le bassin de la Mékrou, neuf (9) modèles hydrologiques pluie-débit globaux dont un à base physique et huit (8) conceptuels ont été testés pour reproduire les écoulements à l’exutoire de Kompongou. Il s’agit des modèles GR4J, AWBM, TANK, Sacramento, HBV, SMAR, IHACRES, SimHyd et ModHyPMA (modèle à base physique). Quatre (4) critères (le Critère de Nash, le coefficient de détermination, l’écart quadratique moyen et l’écart moyen absolu) ont été utilisés pour quantifier la performance des simulations. Les résultats obtenus montrent la supériorité de ModHyPMA, AWBM et HBV sur les autres modèles. Malgré que les modèles GR4J, SimHyd et TANK aient des performances en dessous de ceux de ModHyPMA, AWBM et HBV, ces modèles ont également montré de grande qualité de robustesse à simuler les écoulements à l’exutoire du bassin. Par contre les modèles SMAR, IHACRES, et Sacramento ont donné de faibles résultats surtout en validation pour les deux derniers modèles. L’approche multi-modèle développée à partir de plusieurs combinaisons de cinq modèles a permis une amélioration considérable des performances des modèles en calage et en validation.
Keywords: modélisation pluie-débit; modèles conceptuels; modèles à base physique; multi-modèles.

Simulation and Experimental Study on Low Emission Technology using Multi-Cylinder Turbo Charged Precooled Compression Ignition Engine
425-435
S.Abishek and M.Velliangiri
Abstract:
This research article mainly focuses on Exhaust Gas Recirculation (EGR), influence of injection pressure, injection duration, performance characteristics and emissions on multi-cylinder turbo charged precooled compression ignition engine. The experiment is conducted with two different nozzle diameters of 0.2mm and 0.234mm. Thermodynamic simulation tool Diesel-RK is used for combustion simulation with various load conditions and constant engine speed of 1500rpm. The simulation results are compared for various mass flow rates of exhaust gas recirculation. The combustion simulation with EGR mass flow rates is varied from 0 to 45%. The exhaust gas emission, NOx is very low using EGR system. The Specific Fuel Consumption (SFC) is slightly increased with EGR mode. EGR system is used for achieving ultra-low NOx emissions in multi cylinder turbocharged pre cooled compression ignition engine. The SFC levels slightly decrease with split injection. The NOx levels are found to be lower in split injection compared to single injection. EGR is optimized at 30% and NOx level attained experimentally is 0.36g/kWh at this level agrees better with simulated results.
Keywords: Ultra low NOx, Exhaust Gas Recirculation, Specific fuel consumption, Emission norms, Thermodynamic simulation.

Late Payments in Commercial Transactions in the European Union: Are we Getting Better?
436-447
Jovan Shopovski
Abstract:
Over the years, late payments in commercial transactions have been recognized as one of the main problems of the EU companies, especially for Small and Medium sized (SMEs) ones. The European Union has made several efforts to solve this problem through legal regulations, which are supposed to accelerate the payment processes between companies.
This descriptive paper is aimed at analyzing the current situation in the area of late payments in commercial transactions in the European Union. In the first part of this paper, an overview of the legal institutes, which regulate the area of late payments at the EU level, will be given.
In the second part, the payment behaviour across the EU Member States will be analyzed by presenting data from payment reports of 2005, 2014, and 2015. Moreover, data on payment duration and payment delays of geographically different EU Member States (Germany, Hungary, France, and Italy) will be presented. The latter is carried out in order to compare the payment behaviour in three different periods and stakeholders (Consumer to Business, Business to Business, or Business to Public Sector).
The final goal of this study on the one hand is to arrive at a conclusion of whether current legal provisions, in combating late payments in the EU, are providing the expected results and on the other hand to locate the main problems in this area and suggest further actions.
Keywords: Late payments, European Union, Companies, Directives, Legislation

Characteristics of GCC Islamic Banks Investment in Malaysia
448-457
Ali Abusalah Elmabrok Mohammed and Ng Kim-Soon
Abstract:
Islamic banking and finance services between the GCC countries and Malaysia have entered into an optimistic phase. With the growing demand for the Islamic financial services from Muslim investors, traders as well as consumers, there are now numerous banking and financial institutions from the GCC countries investments in Malaysia. This is due to its popularity of being the Islamic financial hub in the South East Asia. It is the main target for GCC Islamic banks for providing Islamic finance services to the region. This study investigates the characteristics of GCC Islamic banks that have invested in Malaysia. Data was obtained from different research strands: prior documentation and other related databases, Internet searches, websites, libraries; progress reports. Semi-structured interviews and literature survey and problem definition have been done. The dataset consists of all GCC Islamic banks listed in Bank Negara Malaysia during the period 2005-2015. This study concluded that there are 6 GCC Islamic banks that have investments in Malaysia, these banks are named Al Rajhi bank, Kuwait Finance house, Asian Finance bank, Alkhair International Islamic bank, Elaf bank and RUSD Investment bank. They are all Islamic banks and one of them, Elaf bank was acquired by another bank and was removed from the list of licensed banking institutions in Malaysia.
Keywords: Characteristics, Islamic banking and finance, Investment, Gulf Cooperation Council and Malaysia.

Model of Secure Device Interaction Protocol in Mobile and Sensor Networks
458-463
Ali Jaber Tayh
Abstract:
Mobile devices are becoming truly ubiquitous, and mobile computing is making revolutionary changes in the computer world. Mobile devices can constitute heterogeneous networks that provide access to information in online mode from any point in the world. This leads to the issue of ensuring mobile network and information system security, which is quite a complex but relevant task. Such a security system should be easy to use and manage, and also ensure higher reliability. This article describes an approach to designing a security system and interaction protocols for protection of mobile and sensor networks by means of using cryptographic algorithms. In this work, a model of a distributed protocol for authentication in the wireless sensor networks is offered to enable devices to carry out the authentication procedure without using a conventional centralized authentication server
Keywords: