• Journal of the Korea Convergence Society
• /
• v.10 no.10
• /
• pp.7-13
• /
• 2019

The natural gas (NG), mostly methane leaks into the air, it is a big problem for the climate. detected NG leaks under U.S. city streets and collected data. In this paper, we introduced a Deep Neural Network (DNN) classification of prediction for a level of NS leak. The proposed method is OrdinalEncoder(OE) based K-means clustering and Multilayer Perceptron(MLP) for predicting NG leak. The 15 features are the input neurons and the using backpropagation. In this paper, we propose the OE method for labeling target data using k-means clustering and compared normalization methods performance for NG leak prediction. There five normalization methods used. We have shown that our proposed OE based MLP method is accuracy 97.7%, F1-score 96.4%, which is relatively higher than the other methods. The system has implemented SPSS and Python, including its performance, is tested on real open data.

• Journal of Intelligence and Information Systems
• /
• v.22 no.2
• /
• pp.127-142
• /
• 2016

Deep learning model is a kind of neural networks that allows multiple hidden layers. There are various deep learning architectures such as convolutional neural networks, deep belief networks and recurrent neural networks. Those have been applied to fields like computer vision, automatic speech recognition, natural language processing, audio recognition and bioinformatics where they have been shown to produce state-of-the-art results on various tasks. Among those architectures, convolutional neural networks and recurrent neural networks are classified as the supervised learning model. And in recent years, those supervised learning models have gained more popularity than unsupervised learning models such as deep belief networks, because supervised learning models have shown fashionable applications in such fields mentioned above. Deep learning models can be trained with backpropagation algorithm. Backpropagation is an abbreviation for "backward propagation of errors" and a common method of training artificial neural networks used in conjunction with an optimization method such as gradient descent. The method calculates the gradient of an error function with respect to all the weights in the network. The gradient is fed to the optimization method which in turn uses it to update the weights, in an attempt to minimize the error function. Convolutional neural networks use a special architecture which is particularly well-adapted to classify images. Using this architecture makes convolutional networks fast to train. This, in turn, helps us train deep, muti-layer networks, which are very good at classifying images. These days, deep convolutional networks are used in most neural networks for image recognition. Convolutional neural networks use three basic ideas: local receptive fields, shared weights, and pooling. By local receptive fields, we mean that each neuron in the first(or any) hidden layer will be connected to a small region of the input(or previous layer's) neurons. Shared weights mean that we're going to use the same weights and bias for each of the local receptive field. This means that all the neurons in the hidden layer detect exactly the same feature, just at different locations in the input image. In addition to the convolutional layers just described, convolutional neural networks also contain pooling layers. Pooling layers are usually used immediately after convolutional layers. What the pooling layers do is to simplify the information in the output from the convolutional layer. Recent convolutional network architectures have 10 to 20 hidden layers and billions of connections between units. Training deep learning networks has taken weeks several years ago, but thanks to progress in GPU and algorithm enhancement, training time has reduced to several hours. Neural networks with time-varying behavior are known as recurrent neural networks or RNNs. A recurrent neural network is a class of artificial neural network where connections between units form a directed cycle. This creates an internal state of the network which allows it to exhibit dynamic temporal behavior. Unlike feedforward neural networks, RNNs can use their internal memory to process arbitrary sequences of inputs. Early RNN models turned out to be very difficult to train, harder even than deep feedforward networks. The reason is the unstable gradient problem such as vanishing gradient and exploding gradient. The gradient can get smaller and smaller as it is propagated back through layers. This makes learning in early layers extremely slow. The problem actually gets worse in RNNs, since gradients aren't just propagated backward through layers, they're propagated backward through time. If the network runs for a long time, that can make the gradient extremely unstable and hard to learn from. It has been possible to incorporate an idea known as long short-term memory units (LSTMs) into RNNs. LSTMs make it much easier to get good results when training RNNs, and many recent papers make use of LSTMs or related ideas.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.4C
• /
• pp.351-363
• /
• 2006

The 8-VSB modulation, the transmission standard for the terrestrial digital television(DTV) of the Advanced Television Systems Committee(ATSC), suffers from multipath fading because it conveys information on the amplitude. To solve this problem, decision feedback equalizers(DFE's) have been commonly used in terrestrial DTV receivers. However, under severe channels, such as a 0 dB ghost channel or a single frequency network (SFN) channel, the DFE shows unstable convergence due to the error propagation caused by decision errors. Instead of unstable time-domain DFE schemes, by proposing a frequency-domain direct-inversion equalization method, we try to guarantee stable equalization and achieve low symbol error rates. To secure the existence of a channel inverse, channel-matched filtering and noncausal filtering are carried out prior to equalization. Simulation results show that the proposed method performs much better than existing DFE schemes in terms of both the stability and the symbol error rate.

• Journal of KIISE:Computing Practices and Letters
• /
• v.8 no.3
• /
• pp.344-355
• /
• 2002

Having a variety of good characteristics against other pattern recognition techniques, Multilayer Perceptron (MLP) has been used in wide applications. But, it is known that Error Backpropagation (EBP) algorithm which MLP uses in learning has a defect that requires relatively long leaning time. Because learning data in pattern recognition contain abundant redundancies, in order to increase learning speed it is very effective to use online-based teaming methods, which update parameters of MLP pattern by pattern. Typical online EBP algorithm applies fixed learning rate for each update of parameters. Though a large amount of speedup with online EBP can be obtained by choosing an appropriate fixed rate, fixing the rate leads to the problem that the algorithm cannot respond effectively to different leaning phases as the phases change and the learning pattern areas vary. To solve this problem, this paper defines learning as three phases and proposes a Instant Learning by Varying Rate and Skipping (ILVRS) method to reflect only necessary patterns when learning phases change. The basic concept of ILVRS is as follows. To discriminate and use necessary patterns which change as learning proceeds, (1) ILVRS uses a variable learning rate which is an error calculated from each pattern and is suppressed within a proper range, and (2) ILVRS bypasses unnecessary patterns in loaming phases. In this paper, an experimentation is conducted for speaker verification as an application of pattern recognition, and the results are presented to verify the performance of ILVRS.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.6B
• /
• pp.597-603
• /
• 2006

The purpose of this study is to improve the short term rainfall forecast skill using neural network model that can deal with the non-linear behavior between satellite data and ground observation, and minimize the flood damage. To overcome the geographical limitation of Korean peninsula and get the long forecast lead time of 3 to 6 hour, the developed rainfall forecast model took satellite imageries and wide range AWS data. The architecture of neural network model is a multi-layer neural network which consists of one input layer, one hidden layer, and one output layer. Neural network is trained using a momentum back propagation algorithm. Flood was estimated using rainfall forecasts. We developed a dynamic flood inundation model which is associated with 1-dimensional flood routing model. Therefore the model can forecast flood aspect in a protected lowland by levee failure of river. In the case of multiple levee breaks at main stream and tributaries, the developed flood inundation model can estimate flood level in a river and inundation level and area in a protected lowland simultaneously.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.39 no.3
• /
• pp.18-31
• /
• 2002

In this study, we introduce a concept of advanced neurofuzzy polynomial networks(ANFPN), a hybrid modeling architecture combining neurofuzzy networks(NFN) and polynomial neural networks(PNN). These networks are highly nonlinear rule-based models. The development of the ANFPN dwells on the technologies of Computational Intelligence(Cl), namely fuzzy sets, neural networks and genetic algorithms. NFN contributes to the formation of the premise part of the rule-based structure of the ANFPN. The consequence part of the ANFPN is designed using PNN. At the premise part of the ANFPN, NFN uses both the simplified fuzzy inference and error back-propagation learning rule. The parameters of the membership functions, learning rates and momentum coefficients are adjusted with the use of genetic optimization. As the consequence structure of ANFPN, PNN is a flexible network architecture whose structure(topology) is developed through learning. In particular, the number of layers and nodes of the PNN are not fixed in advance but is generated in a dynamic way. In this study, we introduce two kinds of ANFPN architectures, namely the basic and the modified one. Here the basic and the modified architecture depend on the number of input variables and the order of polynomial in each layer of PNN structure. Owing to the specific features of two combined architectures, it is possible to consider the nonlinear characteristics of process system and to obtain the better output performance with superb predictive ability. The availability and feasibility of the ANFPN are discussed and illustrated with the aid of two representative numerical examples. The results show that the proposed ANFPN can produce the model with higher accuracy and predictive ability than any other method presented previously.

• Journal of the Korean Geotechnical Society
• /
• v.20 no.8
• /
• pp.67-75
• /
• 2004

The landslide is one of the most significant natural disasters, which cause a lot of loss of human lives and properties. The landslides in natural slopes generally occur by complicated problems such as soil properties, topography, and geology. Artificial Neural Network (ANN) model is efficient computing technique that is widely used to solve complicated problems in many research fields. In this paper, the ANN model with application of error back propagation method was proposed for estimation of landslide hazard in natural slope. This model can evaluate the possibility of landslide hazard with two different approaches: one considering only soil properties; the other considering soil properties, topography, and geology. In order to evaluate reasonably the landslide hazard, the SlideEval (Ver, 1.0) program was developed using the ANN model. The evaluation of slope stability using the ANN model shows a high accuracy. Especially, the prediction of landslides using the ANN model gives more stable and accurate results in the case of considering such factors as soil, topographic and geological properties together. As a result of comparison with the statistical analysis(Korea Institute of Geosciences and Mineral Resources, 2003), the analysis using the ANN model is approximately equal to the statistical analysis. Therefore, the SlideEval (Ver. 1.0) program using ANN model can predict landslides hazard and estimate the slope stability.

• Korean Journal of Cognitive Science
• /
• v.7 no.1
• /
• pp.57-74
• /
• 1996

This paper presents a comparison of recognition performance of the features used inthe recent handwritten korean character recognition.This research aims at providing the basis for feature selecion in order to improve not only the recognition rate but also the efficiency of recognition system.For the comparison of feature performace,we analyzed the characteristics of theose features and then,classified them into three rypes:global feature(image transformation)type,statistical feature type,and local/ topological feature type.For each type,we selected four or five features which seem more suitable to represent the characteristics of korean alphabet,and performed recongition experiments for the first consonant,horizontal vowel,and vertical vowel of a korean character, respectively.The classifier used in our experiments is a multi-layered perceptron with one hidden layer which is trained with backpropagation algorithm.The training and test data in the experiment are taken from 30sets of PE92. Experimental results show that 1)local/topological features outperform the other two type features in terms of recognition rates 2)mesh and projection features in statical feature type,walsh and DCT features in global feature type,and gradient and concavity features in local/topological feature type outperform the others in each type, respectively.

• Journal of the Institute of Convergence Signal Processing
• /
• v.10 no.1
• /
• pp.7-12
• /
• 2009

In this paper, we propose a directional feature extraction method for off-line handwritten numerals by using the morphological operations. Direction features are obtained from four directional line images, each of which contains horizontal, vertical, right-diagonal and left-diagonal lines in entire numeral lines. Conventional method for extracting directional features uses Kirsch masks which generate edge-shaped double line images for each direction, whereas our method uses directional erosion operations and generate single line images for each direction. To apply these directional erosion operations to the numeral image, preprocessing steps such as thinning and dilation are required, but resultant directional lines are more similar to numeral lines themselves. Our four [$4{\times}4$] directional features of a numeral are obtained from four directional line images through a zoning method. For obtaining the higher recognition rates of the handwrittern numerals, we use the multiple feature which is comprised of our proposed feature and the conventional features of a kirsch directional feature and a concavity feature. For recognition test with given features, we use a multi-layer perceptron neural network classifier which is trained with the back propagation algorithm. Through the experiments with the CENPARMI numeral database of Concordia University, we have achieved a recognition rate of 98.35%.

• The KIPS Transactions:PartB
• /
• v.15B no.1
• /
• pp.45-52
• /
• 2008

As the recent development of the IT services, there is a urgent need of effective diagnosis system to present appropriate solution for the complicated problems of breakdown control, a cause analysis of breakdown and others. So we propose an intelligent diagnosis system that integrates the case-based reasoning and the artificial neural network to improve the system performance and to achieve optimal diagnosis. The case-based reasoning is a reasoning method that resolves the problems presented in current time through the past cases (experience). And it enables to make efficient reasoning by means of less complicated knowledge acquisition process, especially in the domain where it is difficult to extract formal rules. However, reasoning by using the case-based reasoning alone in diagnosis problem domain causes a problem of suggesting multiple causes on a given symptom. Since the suggested multiple causes of given symptom has the same weight, the unnecessary causes are also examined as well. In order to resolve such problems, the back-propagation learning algorithm of the artificial neural network is used to train the pairs of the causes and associated symptoms and find out the cause with the highest weight for occurrence to make more clarified and reliable diagnosis.