Artificial neural networks are systems composed of interconnected simple computing units known as artificial neurons which simulate some properties of their biological counterparts. They have been developed and studied for understanding how brains function, and for computational purposes. Two kinds of architecture of Neural Network Models(NNMs) are the most popular, Recurrent and Feed-forward. ...

DoS attack tools have become increasingly sophisticated challenging the existing detection systems to continually improve their performances. In this paper we present a victimend DoS detection method based on Artificial Neural Networks (ANN). In the proposed method a Feed-forward Neural Network (FNN) is optimized to accurately detect DoS attack with minimum resources usage. The proposed method ...

A new artificial neural network based on decision-making approach for water quality management to control environmental pollution is presented. Previous research on water quality management problems has shown that traditional optimization techniques and an expert-system approach do not provide an educated solution comparing with decision making approach, which is related to the interpretation o...

Predicting Tra c flow in the busiest cities has become a popular research area in the past decades. The rapid development of intelligent tra c management system attracts the software industry to come up with e cient tools for tra c prediction over the roads. In this study, Discrete Wavelet Transformation (DWT) is employed with Artificial Neural Network (ANN) to forecast the tra c flow over the ...

We examine various and different approaches for the prediction of economic crisis periods of US economy. We examine the traditional econometric discrete choice Logit and Probit models then a feed-forward neural network (FFNN) model and finally we apply an Adaptive Neuro-Fuzzy Inference System (ANFIS). We examine the period 1950-2009, where we take as the in-sample or training period 1950-2005, ...

Artificial neural networks can be trained with relatively low-precision floating-point and fixed-point arithmetic, using between one and 16 bits. Previous works have focused on relatively wide-but-shallow, feed-forward networks. We introduce a quantization scheme that is compatible with training very deep neural networks. Quantizing the network activations in the middle of each batch-normalizat...