BPNN

AcronymDefinition
BPNNBack Propagation Neural Network
BPNNBusiness Partner Network Number
References in periodicals archive ?
The performance analysis of SVM, k-NN and BPNN classifiers using DCT, DWT and combined DCT-DWT features is given in Table.
Finally, the results of the proposed fault diagnosis model based on BRNN are compared with those of BPNN, RBF and GRNN, as shown in Table 5.
Neural Network Simulator "BrainMaker"(California Scientific Software 1996) developed a three-layer BPNN model was used for this program.
Further analysis of results led to the conclusion that the best results were obtained by the Sigmoid transfer function for the BPNN.
1) Network training: The network training employs BPNN algorithm by assigning the values of Y, M, and C from the transformation YMC value database as input value and that of X, Y, Z as output.
In a follow-up study that expanded the number of input variables to 18, Johansson and Sonstrod used a BPNN on 100 races at Gulf Greyhound Park and found 24.
Comparison of the Proposed AFDD Strategy with BPNN Classified as True Fault Classifiers Complk Condflt Evapflt Llrestr Refund SVM 97% 2% 1% 0 0 Complk BP 82% 9% 5% 2% 1% SVM 2% 95% 1% 2% 0 Condflt BP 7% 84% 2% 5% 1% SVM 2% 1% 94% 3% 0 Evapflt BP 4% 2% 84% 9% 0 SVM 2% 2% 3% 93% 0 Llrestr BP 3% 5% 7% 82% 1% SVM 0 0 0 0 98% Refund BP 1% 1% 0 2% 87% SVM 0 0 0 0 0 Refovr BP 0 1% 2% 1% 1% SVM 0 0 0 0 0 TXVund BP 2% 1% 1% 4% 0 SVM 0 0 0 0 1% TXVovr BP 1% 1% 3% 3% 0 Classified as True Fault Classifiers Refovr TXVund TXVovr SVM 0 0 0 Complk BP 0 1% 0 SVM 0 0 0 Condflt BP 0 1% 0 SVM 0 0 0 Evapflt BP 0 1% 0 SVM 0 0 0 Llrestr BP 0 1% 1% SVM 0 1% 1% Refund BP 0 5% 4% SVM 100% 0 0 Refovr BP 93% 0 2% SVM 0 100% 0 TXVund BP 0 92% 0 SVM 0 0 99% TXVovr BP 1% 0 91%
BPNN can be trained using the historical data of a time series in order to capture the non-linear characteristics of the specific time series.
The proposed method consists of four main steps: Noise elimination, Segmentation, Pixel Value Matching (PVM) and BPNN classification.
According to the actual application, providing that both the input and output number of node and the input and output values in BPNN have been confirmed, activation function adopts S type function.
The first type of neural network to which the OCR problem was tested for was BPNN which is a feed-forward network.
13) introduced the fuzzy rule into the BPNN model for predicting the flank wear of drill bit.