European Journal of Scientific Research
Volume 121 No 4
Experimental Invesitigation of Training Algorithms used in Backpropagation Artificial Neural Networks to Apply Curve Fitting
Abdullah Al-Hasanat, Haitham Alasha'ary, Khaled Matrouk, Ziad Al-Qadi and Hasan Al-Shalabi
Several backpropagation neural network training methods were tested for curve fitting, it was shown that Levenberg-Marquardt method of training is the most accurate, and is the most suitable method which can be used to minimize the mean square error and the training time. It was shown also that Levenberg-Marquardt training method acheives the best performance in sipte of curve complexity.
Keywords: BPNN, Training Method, MSE, Training Time.
Effet du Cultivar, du Prétraitement et de la Technique de Séchage sur Quelques Paramètres Physico-Chimiques des Figues Séchées de Sept Cultivars Locaux du Figuier (Ficus Carica L.) au Maroc
Ait Haddou L, Blenzar A, Messaoudi Z, Van Damme P, Boutkhil S and Et Boukdame A
Le séchage des figues (fruits du Ficus carica L. Rosaceae), au Maroc, est généralement réalisé de façon traditionnelle sans aucun traitement préalable. Avec une telle façon longue on peut craindre la réhydratation nocturne, des contaminations par l’Aspergillus flavus et la détérioration des fruits: ceci emmène à un produit fini de qualité médiocre. Ces problèmes peuvent être surmontés par l’utilisation de certains prétraitements physico-chimiques et par l’amélioration de la méthode de séchage traditionnelle. Pour améliorer les conditions et réduire la durée de séchage; des tests sur l’effet du cultivar, de certains prétraitements et de la technique adoptée sur la durée du séchage et la qualité physico-chimique des figues séchées des cultivars : "Hzzat", "Ournaksi", "Lemtel", "El Beida", "Khoubzi", "Sebti" et "Nabout" ont été menés. En effet, les fruits mûrs de chaque cultivar ont été traités par un trempage alcalin plus une sulfitation ou uniquement par sulfitation; et ils ont en suite subi une déshydratation sous tunnel et à l’air libre.
Les résultats obtenus ont montré l’effet du cultivar sur la durée de séchage, le diamètre du fruit, la teneur en matière sèche, en matière minérale, en matière organique, en fibres totaux, en protéines totales, en glucose, en fructose, en potassium, en calcium et en magnésium. Le cultivar "Ournaksi" a nécessité moins du temps de séchage et possède plus de potassium et du fructose; le diamètre du cultivar "El Beida" et sa teneur en matière sèche, en glucose, en calcium et en magnésium sont les plus élevés. "Khoubzi" possède une teneur en fibres et en protéines plus importante et a aussi un diamètre assez élevé. "Lemtel" a détenu aussi des valeurs assez importantes en potassium et en magnésium.
Le traitement et la technique ont un effet uniquement sur la durée de séchage. En effet, l’analyse des résultats a montrée que le traitement alcalin a permis un gain de 12 à 32% sur la durée de séchage sous tunnel selon le cultivar. Le minimum et le maximum ont étés enregistrés, respectivement, pour les cultivars "Nabout" et "Khoubzi". Pour le séchage à l’air libre, le gain sur la durée de séchage par le traitement est de 8 à 18% selon le cultivar. Le minimum et le maximum ont étés enregistrés, respectivement, pour les cultivars "Khoubzi" et "Hzzat". Le séchage sous tunnel est plus rapide qu’à l’air libre et a permis d’économiser de 30 à 64% en temps de séchage, selon le cultivar, par rapport au séchage à l’air libre. Le minimum et le maximum ont étés enregistrés, respectivement, pour les cultivars "Nabout" et "Khoubzi".
Motsclés: Traitement alcalin, sulfitation, tunnel, air libre, durée de séchage, qualité organoleptique.
Discounted Semi-Markov Decision Process Based Structural Health Monitoring Model for Civil Infrastructure Systems using Wireless Smart Sensor Networks
Abayomi M. Ajofoyinbo and David O. Olowokere
This paper presents a discounted semi-Markov decision process based Structural Health Monitoring (SHM) model for Civil Infrastructure Systems (CIS) using wireless smart sensor networks. A wireless smart sensor node typically consists of the power, processing, communication (radio) and the physical sensors subsystems. The communication subsystem consumes more energy than the other subsystems. In spite of the advances in Wireless Smart Sensor Networks (WSSN), power consumption remains a challenge that must be addressed for the wireless smart sensors to be more widely adopted in SHM applications. In this paper, the operation of wireless smart sensor networks is characterized as Semi-Markov Decision Process (SMDP) consisting of four states, namely: sleep, sense, process/store, and transmit/receive.
A bridge structure is considered wherein wireless smart sensors deployment follows linear topology. The objective is to choose policy that offers optimal energy-efficient transmission/reception of data packets for measured vibration-based dynamic response of civil structure. We formulate relevant mathematical representations for this SHM model and test its efficacy via numerical analysis, which is implemented in MATLAB software. It is shown by the results obtained that this model can increase energy efficiency in SHM applications using wireless smart sensor networks.
Keywords: Structural health monitoring, energy-efficient, semi-Markov decision process, wireless smart sensors, civil structure.
Applying Classification Algorithms to Measure Students’ Interactivity with Moodle
Asmaa G. Hemdan, Essam Eldean F. Elfharkrny and Safaa Sayed Mahmoud
This paper presents an approach to classify students in order to predict their interactivity level on Moodle LMS based on features extracted from Moodle MySql database. The comparative study was applied with a total of 89 students using different courses on LMS in order to determine the best classifier to be used in measuring students’ interactivity with Moodle from their past behavior on the system. Classification techniques were applied from different classifiers families such as Naïve Bayes from Bayesian Network, J48 as a decision tree, JRip Rule based algorithm and kNN lazy classifier. In this study, the Decision Tree was more appropriate than other algorithms, while Rule-based algorithm replaces the second algorithm in measured parameters results followed by Bayesian Network and finally lazy classifiers. The results of this case study give insight into techniques for appropriately measuring students’ interactivity with LMS and demonstrate the maturity of multiple open source tools.
Keywords: E-learning, Educational data mining, Classification, Naïve Bayes, J48, JRIP, KNN.
Water Vapour Absorbtion by LiBr-H2O Solution
Bayu Rudiyanto, Teti E. Nababan, Armansyah H. Tambunan and Tsair-Wang Chung
In absorption refrigeration system, LiBr as absorbent filters on components of the absorber will absorb water vapor with lower pressure than the evaporator pressure. The rate of absorption of water vapor is a function of concentration, temperature, humidity and pressure of water vapor. The aim of this research is to present to study the effects of concentration, temperature and water vapor pressure to water vapor absorption by LiBr-H2O solution, to determine equilibrium concentration of LiBr-H2O solution, and to study sorption isotherm model. Determination of equilibrium concentration is a function of temperature and humidity obtained : Ce at 40°C and RH 60% was found 37%; at 40°C and RH 70% was 34%; at 40°C and RH 80% was 29%; at 45°C and RH 60% was 37.5%; at 45°C and RH 70% was 36%; and at 45°C and RH 80% was 34.5%. The calculation results of Qe using Qe sorption isotherm model on temperature 40°C and 45°C were not to different. The calculation results of the percentage errors between Qe data and model from each of the humidity, obtained the lowest percentage of errors in model Freundlich.
O solution, Equilibrium concentration, Sorption isotherm model, Model Freundlich
Prevalence and Risk Factor of Soil-Transmitted Helminths Infections among Ethnic Minority Groups in Primary School Students, Nan Province, Thailand
Katekaew Seangpraw, Surasak Taneepanichskul and Ratana Somrongthrong
Soil-Transmitted Helminthes (STH) it one of the major health concerns in rural area in Thailand. Nan province, there are lacks of survey data related to these infections in the study area, particularly among indigenous communities people in primary students, whose inhabitants have plenty of access to health care services, including good dwelling and adequate sanitary conditions.
Objective: (i) To assess the prevalence of Soil-Transmitted Helminth infections among primary students, (ii) to identify risk factors associated with local infection, and (iii) to assess the primary students’ knowledge, attitude, and preventive behavior toward STH infections.
A cross-sectional study was carried in primary school in Bo Kluea district, Nan province in 2014. A total of 132 (males =68, females =64) students (ages 10-12 years) were face to face interviews by trained researchers. In addition, stool examinations using Kato's thick smear technique were conducted. The results were presented as descriptive statistics and inferential statistical including Chi Square. Set the level of significant at p-value <0.05.
The prevalence of STH found to be 14.4 % (19/132). Ascaris lumbricoides
infection was the most prominent (10.6%), followed by Hookworm (1.6%), and Trichuris trichiura (0.8%),
respectively. Infection was prevalent in males (57.9%) more than in females (42.1%), ages 10-12 years. Prevalence of ethnic was Lua (57.9%), Indigenous, and MlaBri (26.3%, 15.8%) respectively. Most of students never had their stools examination for helminthes (62.9%). Most of them (86.4%) had low f knowledge towards STH and attitude towards the preventive of STH were (89.4%) respectively. Multivariate analysis showed that variables such as household (lack of toilet (p <0.000), father education (p <0.001), income of family (p < 0.000), type of housing (p <0.000), hands after defecation (p <0.021), and wearing shoes (p <0.000) were the risk factors. Moderate level of prevention behavior was associated with significant STH infections (p <0.003).
The findings revealed that most of students had low knowledge and attitude towards STH prevention and control. Preventive measures should address important risk factors. Access to public health care and health education regarding to personal hygiene such as washing hand after defecation, wearing shoes, and proper use of latrines are essential elements. Therefore, multiple intervention strategies should be implemented by using community participation among school students, teachers, parents and community for behavioral changed and promoting sanitary latrine used for reducing STH infection.
Keywords: Soil-Transmitted Helminths, Prevalence, Risk Factor, Ethnic Minority Group, Primary Student.
Analysis of Limiting Factors for Life Cycle Assessment of Biomass Energy Systems
The question is whether biomass energy development can meet rising global electricity demand amid international concerns over fossil fuel dependence, global warming, and land use conflicts. A causal loop diagram illustrates the interrelationships between factors that positively and negatively influence the development of biomass as a renewable energy fuel. This research presents a life cycle assessment (LCA) of biomass energy systems to analyze some of the limiting factors. Limiting factors such as increased land use, fossil fuel use, and corresponding CO2 emissions influence further international biomass development efforts. The life cycle assessment evaluated alternative processes that might increase efficiency. The LCA revealed that integrating Salix short-rotation forests, biological fertilizers, and integrated gasification technologies into the biomass energy system would reduce fossil fuel use and CO2 emissions by 74 percent and land use by roughly 97 percent. Biomass energy systems can become much more efficient and competitive sources of renewable electricity by implementing Salix, biological fertilizer, and gasification technologies.
Keywords: Bio-energy; Bio-power; Biomass energy analysis; Life cycle analysis (LCA); Salix (willow) production; Electric power; Gasification; Ash recycling; Sludge recycling.
Implementation Policy of Licensing Services in Dairi Regency, North Sumatra Province
The study was conducted in the Dairi Regency Government in North Sumatra Province, the background being that licensing generally faces various problems, namely: the implementation of licensing has not been able to produce good service, as the service is still synonymous with uncertain costs, a local elitist culture that emphasizes the proximity of power and inconsistencies, disparities in competence and facilities, the morale of personnel, the lack of a grand design, and does not apply the principles of good governance in a comprehensive manner. The research problem is formulated as follows: How much influence do policy implementation, coordination, the professionalism of personnel and public participation have on the quality of the business licensing service in the Dairi Regency Government of North Sumatra Province? The goal is to analyze the effect of the implementation of policy, the coordination, and professionalism of personnel and public participation in the business licensing service quality in the Government Regency of Dairi of North Sumatra Province and to find a new idea or a new concept in the management of governance, especially with regard to the licensing services. Research hypothesis: Policy implementation, coordination, the professionalism of personnel and public participation have a big influence on the quality of business licensing services in the Government Regency of Dairi of North Sumatra Province. To prove this hypothesis, explanatory research was carried out to examine the influence between variables that are hypothesized in the form of analytical regression with independent variables and the dependent variable, in order to obtain the relationship or influence of variables. The researcher took a sample of 300 people (n = 300) of the population of 1,185 (N = 1,185) using Slovin’s formula. The technique of collecting data includes library research, field studies involving observation, interviews and questionnaires. The hypothesis was tested using the F test and the t-test with the computer tool of SPSS for Windows 17.0
and the results and conclusions obtained were as follows: (1) The effect of the implementation of the policy on the quality of business licensing services was 30.7%. (2) The influence of coordination on the quality of business licensing services was 71.8%. (3) The influence of the professional personnel on the quality of business licensing services was 63.9%. (4) The influence of public participation on the quality of business licensing services was 30.9%.
Keywords: Business licensing, policy implementation, coordination, professionalism of officers, community participation.
Cascade Neural Network to Control A MobileRobot in Non-Structured Environment
Adnan Manasreh, Ayman AbuBaker, Mohammad H. Alomari and Ali M. Baniyounes
Obstacle avoidance is an important task in the field of robotics, since the goal of autonomous robot is to reach the destination without collision. Several algorithms have been proposed for obstacle avoidance, having drawbacks and benefits. In this paper, cascade neural network is used to tackle the problem of mobile robot autonomous navigation in a non-structured environment. The objective is to make the robot move in unstructured environment until it reaches its target. The proposed approach uses the adaptive cascade neural network instead of fuzzification and inference engines. A comparison is done for the processing time of the adaptive hybrid controller and many other approaches reported in the literature. The proposed neuro-fuzzy controller showed a good response in real-time applications by avoiding all the tested obstacles. The neuro-fuzzy controller can efficiently reduce the processing time by 70% comparing with tradition fuzzy logic controller.
Keywords: Fuzzy Logic, Neuro-Fuzzy, Mobile Robot, Hybrid Controller.
Assessment of Magnesium Content in the Extracts and Manufactured Formulations of the Medicinal Plant Rauwolfia Serpentina
Conjeevaram J. Gunasekar, Imad A. Abu-Yousef, Amin F. Majdalawieh and Srinivasan Narasimhan
has been well known for centuries in Ayurvedic and Siddha medicine for its different pharmacological activities, thus playing a very important role in phytotherapy. Rauwolfia serpentina
extracts are used mainly for treating hypertension. Essential nutrients, which our bodies cannot make on their own, have a great role to, play in treating hypertension. Generally, nutrients don’t work alone, but rather in synergy with other nutrients to bring about some effects. K, Ca and Mg are known to be crucial elements in the treatment of hypertension. Hypertension is greatly related to salt intake. Vegetarian diets tend to lower blood pressure, partially due to the fact that fruits and vegetables contain K, Ca, and Mg; blood pressure lowering minerals. The nervous system contains millions of neurons and supporting cells. Neurons interact with other cell types like muscle cells and they help to stimulate the sense of seeing light, hearing sound, feeling touch, etc. In this study, we were mainly interested in determining the level of Mg in Rauwolfia serpentina
extracts and three of its formulations that are available in the market for their therapeutic use. Our findings reveal that Rauwolfia serpentina
extracts and manufactured formulations contain considerable levels of Mg, which may explain, at least in part, the therapeutic efficacy of Rauwolfia serpentina
in treating various conditions including hypertension, heart diseases, diabetes, and muscle dysfunction.
Keywords: Rauwolfia serpentina
, elemental profile, ICP-AES, anti-hypertension, Mg.