There are many aims of this book: The first aim is to develop a model equation that describes the spread of contamination through soils which can be used to determine the rate of environmental contamination by estimate the concentration of heavy metals (HMs) in soil. The developed model equation can be considered as a good representation for a problem of environmental contamination. The second aim of this work is to design two feed forward neural networks (FFNN) as an alternative accurate technique to determine the rate of environmental contamination which can be used to solve the model equation. The first network is to simulate the soil parameters which can be used as input data in the second suggested network, while the second network simulates to estimate the concentration of heavy metals. The third aim is to develop a conceptual theory of training stage of neural networks from the perspective of functional analysis and optimization methods. Within this formulation, learning means to solve a variational problem by minimizing a performance function associated to the neural network. The choice of the objective functional depends on the particular application. On the other side, we suggest modification of the performance function to improve the generalization of the suggested networks and to treat the early stopping and local minima problems. The fourth aim is to compare the performance of aforementioned algorithms with regard to predicting ability. Then applied the suggested technique to estimate the concentration of heavy metals such as: Copper(Cu), Lead(Pb), Cadmium(Cd), Cobalt(Co), Zinc(Zn) and Nickel(Ni) in Baghdad soils. First, sixty four soil samples were selected from a phytoremediated contaminated site located in some zones in Baghdad city (residential, industrial, commercial, agricultural and main roads). Second, a series of measurements were performed on the soil samples and analyzed measuring of concentrations for heavy metals using devices such as : Atomic Absorption Spectrophotometer (AAS), X-Ray Fluorescence (XRF) and Inductively Coupled Plasma-Mass Spectrometry (ICP- MS) to get initial concentrations for those heavy metals. Third, simulate and train the suggested networks to get the concentration of heavy metals. The performance of the suggested networks was compared with the traditional laboratory inspecting using the training and test data sets. The results of this book show that the suggested networks trained on experimental measurements can be successfully applied to the rapid and accuracy estimation of concentration of heavy metals. Finally, we suggest some methods for the treatment of contaminated soil by using some herbal plants
The technology of reducing dimensions and choosing variables are very important topics in statistical analysis to multivariate. When two or more of the predictor variables are linked in the complete or incomplete regression relationships, a problem of multicollinearity are occurred which consist of the breach of one basic assumptions of the ordinary least squares method with incorrect estimates results.
There are several methods proposed to address this problem, including the partial least squares (PLS), used to reduce dimensional regression analysis. By using linear transformations that convert a set of variables associated with a high link to a set of new independent variables and unr
... Show More
Abstract:
This research aims to compare Bayesian Method and Full Maximum Likelihood to estimate hierarchical Poisson regression model.
The comparison was done by simulation using different sample sizes (n = 30, 60, 120) and different Frequencies (r = 1000, 5000) for the experiments as was the adoption of the Mean Square Error to compare the preference estimation methods and then choose the best way to appreciate model and concluded that hierarchical Poisson regression model that has been appreciated Full Maximum Likelihood Full Maximum Likelihood with sample size (n = 30) is the best to represent the maternal mortality data after it has been reliance value param
... Show More
A condense study was done to compare between the ordinary estimators. In particular the maximum likelihood estimator and the robust estimator, to estimate the parameters of the mixed model of order one, namely ARMA(1,1) model.
Simulation study was done for a varieties the model. using: small, moderate and large sample sizes, were some new results were obtained. MAPE was used as a statistical criterion for comparison.
Abstract
The grey system model GM(1,1) is the model of the prediction of the time series and the basis of the grey theory. This research presents the methods for estimating parameters of the grey model GM(1,1) is the accumulative method (ACC), the exponential method (EXP), modified exponential method (Mod EXP) and the Particle Swarm Optimization method (PSO). These methods were compared based on the Mean square error (MSE) and the Mean Absolute percentage error (MAPE) as a basis comparator and the simulation method was adopted for the best of the four methods, The best method was obtained and then applied to real data. This data represents the consumption rate of two types of oils a he
... Show More
The effect of subinhibitory concentration of Antibiotics on the Adherence of S.aureus (Coagulase Positive Staphylococci), and S.epidermidis (Coagulase negative Staphylococci) and Pseudomonas aeruginosa, Enterobacter cloacae, Citobacter freundi (Gram negative bacteria) was done and the results revealed that Rifampicin was the best antibiotic inhibiting Staphylococci adherence and Vancomycin has less effect on the adherence of Staphylococci, whereas Tetracyclin was the best antibiotic inhbiting Gram negative bacteria adherence and Amikacin has the lest less effect on inhibiting bacterial adherence.
Abstract
The logistic regression model is one of the nonlinear models that aims at obtaining highly efficient capabilities, It also the researcher an idea of the effect of the explanatory variable on the binary response variable. &nb
... Show More
(5)
(1)
