• Journal of Korea Multimedia Society
• /
• v.6 no.7
• /
• pp.1312-1320
• /
• 2003

In this paper, we consider a multilayer neural network, with a single hidden layer. Error backpropagation learning method used widely in multilayer neural networks has a possibility of local minima due to the inadequate weights and the insufficient number of hidden nodes. So we propose a fuzzy supervised learning algorithm by using self-generation that self-generates hidden nodes by the compound fuzzy single layer perceptron and modified ART1. From the input layer to hidden layer, a modified ART1 is used to produce nodes. And winner take-all method is adopted to the connection weight adaptation, so that a stored pattern for some pattern gets updated. The proposed method has applied to the student identification card images. In simulation results, the proposed method reduces a possibility of local minima and improves learning speed and paralysis than the conventional error backpropagation learning algorithm.

• The KIPS Transactions:PartB
• /
• v.9B no.3
• /
• pp.271-276
• /
• 2002

The test statistic in ANOVA tests has a single or doubly noncentral F distribution and the noncentral F distribution is applied to the calculation of the power functions of tests of general linear hypotheses. Although various approximations of noncentral F distribution are suggested, they are troublesome to compute. In this paper, the calculation of noncentral F distribution is applied to the neural network theory, to solve the computation problem. The neural network consists of the multi-layer perceptron structure and learning process has the algorithm of the backpropagation. Using fables and figs, comparisons are made between the results obtained by neural network theory and the Patnaik's values. Regarding of accuracy and calculation, the results by neural network are efficient than the Patnaik's values.

• The KIPS Transactions:PartB
• /
• v.11B no.6
• /
• pp.735-742
• /
• 2004

We present a new use of common graphics hardware to perform a faster artificial neural network. And we examine the use of GPU enhances the time performance of the image processing system using neural network, In the case of parallel computation of multiple input sets, the vector-matrix products become matrix-matrix multiplications. As a result, we can fully utilize the parallelism of GPU. Sigmoid operation and bias term addition are also implemented using pixel shader on GPU. Our preliminary result shows a performance enhancement of about thirty times faster using ATI RADEON 9800 XT board.

• Journal of Korean Institute of Industrial Engineers
• /
• v.28 no.2
• /
• pp.147-161
• /
• 2002

In novelty detection, one attempts to discriminate abnormal patterns from normal ones. Novelty detection is quite difficult since, unlike usual two class classification problems, only normal patterns are available for training. Auto-Associative Multi-Layer Perceptron (AAMLP) has been shown to provide a good performance based upon the property that novel patterns usually have larger auto-associative errors. In this paper, we give a mathematical analysis of 2-layer AAMLP's output characteristics and empirical results of 2-layer and 4-layer AAMLPs. Various activation functions such as linear, saturated linear and sigmoid are compared. The 2-layer AAMLPs cannot identify non-linear boundaries while the 4-layer ones can. When the data distribution is multi-modal, then an ensemble of AAMLPs, each of which is trained with pre-clustered data is required. This paper contributes to understanding of AAMLP networks and leads to practical recommendations regarding its use.

• Journal of Korea Multimedia Society
• /
• v.2 no.4
• /
• pp.490-495
• /
• 1999

In this paper, a method of improving the learning time and convergence rate is proposed to exploit the advantages of artificial neural networks and fuzzy theory to neuron structure. This method is applied to the XOR Problem, n bit parity problem which is used as the benchmark in neural network structure, and recognition of digit image in the vehicle plate image for practical image application. As a result of the experiments, it does not always guarantee the convergence. However, the network showed improved the teaming time and has the high convergence rate. The proposed network can be extended to an arbitrary layer Though a single layer structure Is considered, the proposed method has a capability of high speed 3earning even on large images.

• Journal of the Korea Computer Industry Society
• /
• v.3 no.5
• /
• pp.547-556
• /
• 2002

This paper was proposed a wave path loss prediction algorithm using multilayer perceptron (MLP) model and wave propagation characteristic parameters for Wireless LAN in indoor radio environments. Receiving power was predicted by calculating indoor path loss in a Wireless LAN that has transmission power of 100mW and frequency of 2.4GHz, and was compared with measured. In the result of measurement shows that there is a difference between predicted and measured receiving power which can be reduced by an accurate analysis of the various path loss factors. In order to fix the access point(AP) positions was used the proposed a wave path loss prediction algorithm, and designed the optimum cell for Wireless LAN.

• Korean Journal of Cognitive Science
• /
• v.22 no.4
• /
• pp.429-447
• /
• 2011

In this paper, we propose a modified deterministic Korean dependency parser using a projectivity. The modified parser is improved by finding errors, such as cross dependency, from the original parsing results and correcting them according to the projectivity and head-final principles. Our parser also uses parsing history information in addition to rich features, which only a deterministic algorithm can use. Results on the modified parser for ETRI(2005) corpus, that consists of complex sentences, show that our parser outperforms other parsers.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.24 no.1
• /
• pp.63-70
• /
• 2010

In this paper, we study the structure of neural network for predicting defect size of steam generator tube. After extracting the features from the eddy current testing (ECT) signals, multi-layer neural networks are used to predict the defect size. In order to maximize the prediction performance for the defect size, we should carefully choose the structure of neural networks, especially the number of neurons in the hidden layer. In this paper, it is shown that, for the prediction of defect size, the number of neurons in the hidden layer can be efficiently determined by using cross-validation.

• Journal of KIISE:Computing Practices and Letters
• /
• v.13 no.5
• /
• pp.338-342
• /
• 2007

In this paper we propose a user adaptive post-processing method to improve the accuracy of speaker dependent, isolated word speech recognition, particularly for mobile devices. Our method considers the recognition result of the basic recognizer simply as a high-level speech feature and processes it further for correct recognition result. Our method learns correlation between the output of the basic recognizer and the correct final results and uses it to correct the erroneous output of the basic recognizer. A multi-layer perceptron model is built for each incorrectly recognized word with high frequency. As the result of experiments, we achieved a significant improvement of 41% in recognition accuracy (41% error correction rate).

• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 2005.11a
• /
• pp.158-161
• /
• 2005

In this study, the defect diagnosis of the gas turbine engine was tried using Support Vector Machine(SVM). It is known that SVM can find the optimal solution mathematically through classifying two groups and searching for the Hyperplane of the arbitrary nonlinear boundary. The method for the decision of the gas turbine defect quantitatively was proposed using the Multi Layer SVM for classifying two groups and it was verified that SVM was shown quicker and more reliable diagnostic results than the existing Multi Layer Perceptron(MLP).