Monaural source separation is a challenging issue due to the fact that there is only a single channel available; however, there is an unlimited range of possible solutions. In this paper, a monaural source separation model based hybrid deep learning model, which consists of convolution neural network (CNN), dense neural network (DNN) and recurrent neural network (RNN), will be presented. A trial and error method will be used to optimize the number of layers in the proposed model. Moreover, the effects of the learning rate, optimization algorithms, and the number of epochs on the separation performance will be explored. Our model was evaluated using the MIR-1K dataset for singing voice separation. Moreover, the proposed approach achi

A general mathematical model for a fixed bed immobilized enzyme reactor was developed to simulate the process of diffusion and reaction inside the biocatalyst particle. The modeling and simulation of starch hydrolysis using immobilized α-amylase were used as a model for this study. Corn starch hydrolysis was carried out at a constant pH of 5.5 and temperature of . The substrate flow rate was ranging from 0.2 to 5.0 mL/min, substrate initial concentrations 1 to 100 g/L. α-amylase was immobilized on to calcium alginate hydrogel beads of 2 mm average diameter. In this work Michaelis-Menten kinetics have been considered. The effect of substrate flow rate (i.e., residence time

<span lang="EN-US">Diabetes is one of the deadliest diseases in the world that can lead to stroke, blindness, organ failure, and amputation of lower limbs. Researches state that diabetes can be controlled if it is detected at an early stage. Scientists are becoming more interested in classification algorithms in diagnosing diseases. In this study, we have analyzed the performance of five classification algorithms namely naïve Bayes, support vector machine, multi layer perceptron artificial neural network, decision tree, and random forest using diabetes dataset that contains the information of 2000 female patients. Various metrics were applied in evaluating the performance of the classifiers such as precision, area under the c

An essential element in English as a foreign language (EFL) learning is vocabulary. There is a big emphasis on learning the new words' meaning from the books or inside classrooms. Also, it is a major part of language teaching as well as being fundamental to the learner but there is a big challenge in vocabulary instruction due to the weak confidence by teachers in selecting the suitable practice in teaching vocabulary or they sometimes unable to specify a suitable time for it during the teaching process. The major aim of this study is to investigate the value of posters in vocabulary learning on the 2nd grade students at Halemat Alsaadia High School in Baghdad – Iraq. It hypothesized that there are no statistically significant differences

Computations of the relative permeability curves were made through their representation by two functions for wetting and nonwetting phases. Each function contains one parameter that controls the shape of the relative permeability curves. The values of these parameters are chosen to minimize an objective function, that is represented as a weighted sum of the squared differences between experimentally measured data and the corresponding data calculated by a mathematical model simulating the experiment. These data comprise the pressure drop across core samples and the recovery response of the displacing phase. Two mathematical models are constructed in this study to simulate incompressible, one-dimensional, two-phase flow. The first model d

   A comparison of double informative and non- informative priors assumed for the parameter of Rayleigh distribution is considered. Three different sets of double priors are included, for a single unknown parameter of Rayleigh distribution. We have assumed three double priors: the square root inverted gamma (SRIG) - the natural conjugate family of priors distribution, the square root inverted gamma – the non-informative distribution, and the natural conjugate family of priors - the non-informative distribution as double priors .The data is generating form three cases from Rayleigh distribution for different samples sizes (small, medium, and large). And Bayes estimators for the parameter is derived under a squared erro

This paper presents L₁-adaptive controller for controlling uncertain parameters and time-varying unknown parameters to control the position of a DC servomotor. For the purpose of comparison, the effectiveness of L₁-adaptive controller for position control of studied servomotor has been examined and compared with another adaptive controller; Model Reference Adaptive Controller (MRAC). Robustness of both L₁-adaptive controller and model reference adaptive controller to different input reference signals and different structures of uncertainty were studied. Three different types of input signals are taken into account; ramp, step and sinusoidal. The L₁-adaptive controller ensured uniformly bounded

Deep Learning Techniques For Skull Stripping of Brain MR Images

One of the diseases on a global scale that causes the main reasons of death is lung cancer. It is considered one of the most lethal diseases in life. Early detection and diagnosis are essential for lung cancer and will provide effective therapy and achieve better outcomes for patients; in recent years, algorithms of Deep Learning have demonstrated crucial promise for their use in medical imaging analysis, especially in lung cancer identification. This paper includes a comparison between a number of different Deep Learning techniques-based models using Computed Tomograph image datasets with traditional Convolution Neural Networks and SequeezeNet models using X-ray data for the automated diagnosis of lung cancer. Although the simple details p