Progress in supervised neural networks

Hush, Don; Horne, Bill G.

doi:10.1109/79.180705

Cited by 1,133 publications

(373 citation statements)

References 71 publications

Supporting

Mentioning

342

Contrasting

Unclassified

Order By: Relevance

“…2). The network is trained using a standard backpropagation algorithm [12] with randomly initialized weights. The backpropagation weight update rule, also called generalized delta rule reads as follows: where p is the learning factor (a constant), l j the error of neuron j (the difference between the real output and the teaching input), net j the net input in neuron j, t j the teaching input of neuron j, o i the output of the preceding neuron j, i the index of a predecessor of the current neuron j with link w ij from i to j, j the index of the current neuron, and k the index of a successor to the current neuron j with link w jk from j to k. A new activation a j (t) of neuron j in step t is computed using the sigmoidal function as transfer function f:…”

Section: Classification Using Neural Networkmentioning

confidence: 99%

Automatic pattern recognition in ECG time series

Sternickel

2002

Computer Methods and Programs in Biomedicine

View full text Add to dashboard Cite

Section: Classification Using Neural Networkmentioning

confidence: 99%

Automatic pattern recognition in ECG time series

Sternickel

2002

Computer Methods and Programs in Biomedicine

View full text Add to dashboard Cite

“…Nevertheless, it is very hard to choose the number of hidden layers (30). Most of literatures indicate that one hidden layer is good enough to validate the prediction and maybe the best to decide for all applied feed-forward network design (38). Thus, in this paper one hidden layer was used for modeling ( Figure 2).…”

Section: Ann Modelingmentioning

confidence: 99%

Artificial Neural Network (ANN) Approach for Predicting Cu Concentration in Drinking Water of Chahnimeh1 Reservoir in Sistan-Balochistan, Iran

2013

View full text Add to dashboard Cite

Background: Access to safe drinking water is one of the basic human rights and essential for healthy life. Concerns about the effects of copper on human health have led to numerous guidelines and regulations limiting its concentrations in water. Objectives: The major goal of this study is to demonstrate artificial neural network model of the Chahnimeh1 reservoir water quality (Heavy metal concentration) and show the potential of the ANN for producing models capable of efficient forecasting of Cu concentration. Materials and Methods: Water samples were collected from Chahnimeh1 reservoir which was the most important source of drinking water in Sistan-balochistan and analyzed for physical quality parameters such as: EC (electric conductivity), TDS(total dissolved solids), T(temperature), pH and heavy metal (Cu) concentration using standard methods. In this study, a three-layer artificial neural network (ANN) model was investigated to predict the Cu concentration in the water of Chahnimeh1 reservoir. The input variables are electric conductivity, total dissolved solids, temperature and pH, while the Cu concentration in water is the output. We applied The Levenberg-Marquardt (LM) algorithm to train ANN. Results: According to the ANN outputs, hidden layer with 7 neurons had the best performance for predicting Cu concentration. Evaluation indexes including MSE and R in this article were obtained as 0.00008 and 0.9346; 0.00019 and 0.8612; 0.00014 and 0.9372 for training, validation and testing date sets respectively. Conclusions: As we can see the ANN outputs values are very close to actual Cu concentration, so indicating that predicted values are accurate and the network design is proper and the input variables well suitable for the prediction of Cu concentration.

show abstract

“…The back-propagation error surface usually consists of a large amount of flat regions as well as extremely steep regions (Hush and Horne, 1993), so there is difficulty how to choose an appropriate value of learning rate. As such, the backpropagation algorithm with a variable learning is desirable [1] .…”

Section: Variable Learning Ratementioning

confidence: 99%

“…Learning phase: In this stage, the network weights are randomly generated and the training process will try to adjust the weights so that the actual output comes out closer to the expected output [8] according to the following procedures: One of the patterns to be learned will be put on the input units. Values of the output of hidden layers units and output layer units are calculated.…”

Section: • Execute Phasementioning

confidence: 99%

Change Detection Using Neural Network with Improvement Factor in Satellite Images

Fkirin¹

2009

American Journal of Environmental Sciences

View full text Add to dashboard Cite

Problem statement:The aim of this study is to investigate the applicability of using the neural network techniques in change detection of remotely sensed data. Approach: In addition, the tuning parameters of the network, such as encoding the output classes, adding the momentum term and learning rate, are investigated in order to achieve best network performance. Results: Neural networkbased change detection system in this study is implemented using back propagation-training algorithm. This trained network is designed to be able to detect efficiently any variation between two images and provide adequate information about the type of changes. In an effort to meet these requirements, neural network scheme with improvement factor, leaning rate and momentum term is proposed to monitor environmental changes in Toshka area, Egypt. Two sets of satellite images with different dates are used, the first set contains of two sample satellite images, the second set of images acquired on 1984, 2000 and 2003. Conclusion/Recommendations: Comparing the output of the proposed model with the mostly used change detection techniques; ratio and classification, results show a great potential as the proposed scheme was able to identify not only the changed and non-changed area but also it was capable to identify the nature of these changes.

show abstract

Progress in supervised neural networks

Cited by 1,133 publications

References 71 publications

Automatic pattern recognition in ECG time series

Automatic pattern recognition in ECG time series

Artificial Neural Network (ANN) Approach for Predicting Cu Concentration in Drinking Water of Chahnimeh1 Reservoir in Sistan-Balochistan, Iran

Change Detection Using Neural Network with Improvement Factor in Satellite Images

Contact Info

Product

Resources

About