Find research datasets worth reusing

BRAIN Journal-An Enhancement over Texture Feature Based Multiclass Image Classification under Unknown Noise-Figure 4. Feed Forward Back propagation Neural Network

This BPNN provides a computationally efficient method for changing the weights in feed forward network, with differentiable activation function units, to learn a training set of input-output data. Being a gradient descent method it minimizes the total squared error of the output computed by the net. The aim is to train the network to achieve a balance between the ability to respond correctly to the input patterns that are used for training and the ability to provide good response to the input that are similar. A typical back propagation network of input layer, one hidden layer and output layer is shown in figure 4.

BRAIN Journal-An Enhancement over Texture Feature Based Multiclass Image Classification under Unknown Noise-Figure 6. Performance in experiment 2 for both approaches

Second experiment is performed on noisy data set with 50 images each class for training of feed forward neural network, and 100 images each class are used for testing phase. This experiment is also carried out with two approaches as discussed in first experiment.

BRAIN Journal-Intelligent Continuous Double Auction method For Service Allocation in Cloud Computing-Figure 5. Comparison on successful allocation

Figure 5 illustrates comparison of successful allocation rate between ICDA and other methods. In the proposed method, intelligent allocation and also time consideration enable consumers to acquire more resources before the deadline and as a result, the number of successful allocation is higher than other methods.

BRAIN Journal-An Enhancement over Texture Feature Based Multiclass Image Classification under Unknown Noise-Figure 9. Performance in 2nd Approach for three data set

This work also deals with classification of multi class images under different constraints of data set. The first experiment is carried out on images without noise, second with Gaussian noise and filtered data set in third experiment. Performance of the classifier using statistical texture features for two approaches are presented in the table 3. It is observed that performance n the first experiment is best in the first data set i.e. data set without noise in both the approach, while the performance is decreased if the same images are affected by Gaussian noise. This is because the texture feature of the original images consists Gaussian pattern also. Filtering of the noise from the second data set improves the result. The table also shows that feature extraction using blocking of the image enhance the average classification rate in all the case.

BRAIN Journal-An Enhancement over Texture Feature Based Multiclass Image Classification under Unknown Noise-Figure 5. Performance in experiment 1 for both approaches

The first experiment is carried out on data set containing 16 class and 50 images of each class for training and 100 images of each class for testing. Initially the original images are resized to 128X128 pixels and six texture features are extracted from 800 training images in the first approach. It produces a feature matrix of size 6X800. Similarly 6X1600 feature matrix is produced for testing stage. Then a neural classifier is designed with ten hidden layers and 16 output layers. Feature matrix of training images are used to train the neural network. After that they are tested on feature matrix of test image. In second approach, the original images are divided into 16 blocks of size 32X32 pixel each block. So it produces 96 features (16 blocksX6 features) for each image and feature matrix of 96X800 for training image and 96X1600 feature matrix for testing samples.

BRAIN Journal-An Enhancement over Texture Feature Based Multiclass Image Classification under Unknown Noise-Figure 3. Applying Wiener Filter on noisy image

The Wiener filter is the mean square error-optimal stationary linear filter for images degraded by additive noise and blurring. It removes the additive noise and inverts the blurring simultaneously. The Wiener filtering is a linear estimation of the original image. The approach is based on a stochastic framework.

BRAIN Journal-Intelligent Continuous Double Auction method For Service Allocation in Cloud Computing-Figure 4. Sharing and tradeoff factor effect on resource utilization

In Figure 4 we consider tradeoff and sharing factors in providers. The result illustrates that by using these factors providers improve resource utilization. Higher resource utilization motivates more providers to participate in the cloud and also enables the cloud market to handle more consumers which influences market efficiency.

BRAIN Journal-An Enhancement over Texture Feature Based Multiclass Image Classification under Unknown Noise-Figure 2. Original image affected by Gaussian noise

In this research work we have detected Gaussian noise (fig 2) pattern in our images based on methodology discussed in [18]. Where statistical moments features are extracted from the noise patterns for noise class detection. This experiment detects the Gaussian noise patterns from images. This leads to applying of wiener filter on noisy images, which gives the best noise removal.

BRAIN Journal-Intelligent Continuous Double Auction method For Service Allocation in Cloud Computing-Figure 1. Resource allocation schema in proposed method

We assume that the resources allocation satisfies the following conditions: • The quantity of a resource can be measured in arbitrary units (e.g. 60 units of resource A). • A resource can be divided into an arbitrary fraction (e.g. a resource of 60 units is divided into 20 units for consumer 1 and 40 units for consumer 2). • A resource request of a service can be divided into sub-requests and acquired from multiple providers (e.g. a resource request of 40 units utilized as 10 units from provider 1 and 30 units from provider 2). Figure 1 shows a cloud computing environment with the proposed mechanism.

BRAIN Journal-High Performance Data mining by Genetic Neural Network-Figure 8 . The Comparision of Run Times

That is distinct that dynamic mutation rate or reduction idea for mutation operator is more better of fixed rate. In fact obtain to high accuracy is result of our idea for mutation operator. The number of hidden layer neurone is important problem for NN. The natural selection by GA help finding the number of hidden layer neurone and it progress on duration generations. The structured model of GANN finds better answer than NN but with much run time in simulation. The learning of GA is much better than NN with back propagation because BP is a method based on gradient descend and local optimum is a serious risk for that. We hope that the number of training samples is more accurate without error, the new algorithm is better. Tests show that the combination of genetic algorithms and neural networks to an acceptable level solves the problem of overfitting.

BRAIN Journal-High Performance Data mining by Genetic Neural Network-Figure 7 . Test Accuracy with prograess generation

In the training phase, the neural network weights errors are minimized and network design problem which the objective function to an acceptable level. In test step we have better results because weights of neural network are adjusted by genetic algorithm and back propagation method. Of course achievement to accuracy with 83.5% is reason using of good feature with minimum error.

BRAIN Journal-Isomorphism Between Estes' Stimulus Fluctuation Model and a Physical- Chemical System-Figure 1. Two compartments containing solution separated by a membrane.

In fact, this equation will be found first if one consults physical or chemical textbooks for diffusion. Also one may be able to find already-existing diffusion simulators to see vivid images of the process.

BRAIN Journal-High Performance Data mining by Genetic Neural Network-Figure 6. Training Accuracy with prograess generation

There are many features will reduce the efficiency of the algorithm and its complexity. Among the methods for selecting the appropriate features, the algorithm is a GA. One of the important parameters for testing methods is accuracy rate on progress generation. In fact accuracy is reverse error in algorithm results. As reader can compare the results of our paper with another works. Figure 4 show that accuracy present for Training step. We achieve to best answers of 800 generation to after generation.

BRAIN Journal-High Performance Data mining by Genetic Neural Network-Figure 4. Structural Crossover

Guided crossover operator is based on the two point separation from parents are selected Left and right parts of them are related to each other by the condition to be meaningful With this new child of his parents is that. But a new generation of the random choice to have reached this stage. The crossover rate is fixed for our algorithm.

BRAIN Journal-High Performance Data mining by Genetic Neural Network-Figure 5. Insertion and Deletion Hidden Layer in NN

Change in NN structure is other method that we used to optimization of solution[18]. Insertion a hidden layer caused to mutation operator is much natural. As connection with father and mother nodes is easily[20],[21]. Weights of node and errors automatically calculated. For each stage of the implementation of the mutation operator in genetic algorithms, neural networks, only one of the nodes in the hidden layer is selected and inserted. These layers are inserted on condition that the definition does not harm the network structure and the action is meaningful. As an added layer can adjust the weights and the connection to the parent node of a network layer to be removed.

BRAIN Journal-High Performance Data mining by Genetic Neural Network-Figure 3. The Structure of Neural Network

A neural network (NN), in the case of artificial neurons called artificial neural network (ANN) or simulated neural network (SNN), is an interconnected group of natural or artificial neurons that uses a mathematical or computational model for information processing based on a connectionist approach to computation. In most cases an ANN is an adaptive system that changes its structure based on external or internal information that flows through the network[9]. In more practical terms neural networks are nonlinear statistical data modelling or decision making tools. They can be used to model complex relationships between inputs and outputs or to find patterns in data. Two neurons neural network active in memory (ON or 1) or disable (Off or 0), and each edge (synapses or connections between nodes) is a weight. Edges with positive weight, stimulate or activate next active node, and edges with negative weight, disable or inhibit the next connected node (if it is active) ones.

BRAIN Journal-On the Idea of a New Artificial Intelligence Based Optimization Algorithm Inspired From the Nature of Vortex-Figure 1. Working mechanism of the VOA.

As it can be seen from the algorithm steps, the VOA employs simple equations. It is an advantage that the algorithm can be formed and applied within optimization problems whereas alternative algorithms may contain some complex solution steps (This situation may be also an disadvantage for the VOA when it is applied in more difficult optimization problems but while the world is transformed into a ‘strong simplicity’, the VOA may be a practical solution approach). The working mechanism of the VOA can be visualized briefly as like in Figure 1.

BRAIN Journal-Brain signal analysis using EEG and Entropy to study the effect of physical and mental tasks on cognitive performance-Figure 2. Energy-VAS subjective measures for the participants (n=12) under two conditions (control and exercise involved cognitive task).

As shown in Figure 2, it was found that there was a statistically significant interaction in the percentage of mental fatigue between the condition type and time-on-task factor times (F(6, 22) = 492.19, p < 0.001) as well as there was a significant main effect of time-on-task (time5 to time30) (F(5, 22) = 463.794, p < 0.001). In addition, there was also a significant main effect in the condition type (F (1, 22) = 713.133, p < 0.001) which represented a large effect size. For the physical fatigue subjective measure, there was a significant difference between the two experimental conditions (p < 0.001), and within the subject test times (p < 0.001). However, there was no significant difference between the means of the concentration visual analogue scale for these two experimental conditions (p = 0.057) despite a significant difference (p < 0.001) in the time-on-task repeated measures.

BRAIN Journal-Electrophysiological Neuroimaging using sLORETA Comparing 100 Schizophrenia Patients to 48 Patients with Major Depression -Figure 4. Histogram age distributions of the 100 Schizophrenia patients illustrating clusters of patients at 20, 30, 40, and 50 years old.

The figure below is a histogram distribution of the ages of one hundred Schizophrenia patients in this study. There appears to be a cyclical peak every ten-years cycles at 20, 30, 40, and 50-year-old patients. This may suggest a recent finding that CD8 T cells, which play a pivotal role in mediating long-term immunity to Toxoplasma, are down-regulated in schizophrenia patients (Bhadra et al., 2013). Based on the aforementioned statements, a vaccine for Toxoplasmosis may be difficult to achieve, but the treatment and screening possibilities in patients with psychosis are available today.

BRAIN Journal-Isolating the Norepinephrine Pathway Comparing Lithium in Bipolar Patients to SSRIs in Depressive Patients-Figure 2. Resting state neuroimaging findings illustrating the action of Lithium following whole brain

The axial, saggital, and coronal MRI activation maps illustrate increased delta frequency band neuronal activity in the 46 patients diagnosed with Bipolar Affective Disorder compared to 32 female patients diagnosed with Major Depressive Disorder of Depressive Episode. The Yellow/Orange shades indicate increased neuronal activity in the right Superior Frontal Gyrus (t=0.920, p=0.05060, BA 6, MNI X=20, Y=0, Z=70) and in the right Cingulate Gyrus (t=0.0846, BA 24, MNI X= 5, Y=0, Z=51). Structural anatomy is shown in grey scale (A – anterior; S – superior; P – posterior; L – left; R – right).