Caption: Figure 4: Comparison the classification accuracy and training time of SVM, I-SVM, and MLPNN
The heart of designing an MLPNN
is the training of network for learning the behaviour of input-output patterns.
model, all the input nodes are in one layer and hidden layer is distributed into one or more hidden layers.
In this work the feed forward multilayered perceptron neural network (MLPNN
), has been used.
Multilayer perceptron neural network (MLPNN
) is a widely used neural network structure in antenna applications.
Estudios anteriores han clasificado varias senales EEG a traves de clasificadores llamados Red Neuronal Perceptron Multicapas (MLPNN
) (Dreiseitl & Ohno-Machado, 2002).
In this architecture, feed-forward Multilayer Perceptron Neural Network (MLPNN
) has been used transforming (r, x) input values to ([[phi].sub.r], [[phi].sub.x], [[theta].sub.r], [[theta].sub.x]) output values.
The most commonly used and simplest network architecture called Multilayer Perceptron Neural Network (MLPNN
Multi-Layer Perceptron Neural Network (MLPNN
) also known as multilayer feed-forward neural network was chosen and used in this study.
(2009) studied the use of MUTNN Multiple Temporal Units Neural Network and PENN Parallel Ensemble Neural Network for passenger flow prediction on the railway and considered these two methods to be more exact than the conventional MLPNN
(Multi-Layer Perception Neural Network) one.
achieved a classification accuracy of 88.3% using multilayer perceptron neural network (MLPNN
) classifier using 10 files of the MIT/BIH arrhythmia database.
Table 1: Results obtained from data analysis Sample Intervals Acceleration of Pumps Hz domain Result N/mm 0-2000, Pump #1 2000-4000 8000 and higher Sound Pump #2 0-2000, between 1000 Need to repair but 2000-10000 and 2000 it will work Pump #3 0-2000, 1000 and lower The pump will be 2000-4000 considered defective Table 2: The Results of the proposed method and MLPNN