• Journal of KIISE:Software and Applications
• /
• v.32 no.11
• /
• pp.1021-1028
• /
• 2005

The general solution for classification and regression problems can be found by matching and modifying matrices with the information in real world and then these matrices are teaming in neural networks. This paper treats primary space as a real world, and dual space that Primary space matches matrices using kernel. In practical study, there are two kinds of problems, complete system which can get an answer using inverse matrix and ill-posed system or singular system which cannot get an answer directly from inverse of the given matrix. Further more the problems are often given by the latter condition; therefore, it is necessary to find regularization parameter to change ill-posed or singular problems into complete system. This paper compares each performance under both classification and regression problems among GCV, L-Curve, which are well known for getting regularization parameter, and kernel methods. Both GCV and L-Curve have excellent performance to get regularization parameters, and the performances are similar although they show little bit different results from the different condition of problems. However, these methods are two-step solution because both have to calculate the regularization parameters to solve given problems, and then those problems can be applied to other solving methods. Compared with UV and L-Curve, kernel methods are one-step solution which is simultaneously teaming a regularization parameter within the teaming process of pattern weights. This paper also suggests dynamic momentum which is leaning under the limited proportional condition between learning epoch and the performance of given problems to increase performance and precision for regularization. Finally, this paper shows the results that suggested solution can get better or equivalent results compared with GCV and L-Curve through the experiments using Iris data which are used to consider standard data in classification, Gaussian data which are typical data for singular system, and Shaw data which is an one-dimension image restoration problems.

• Computers and Concrete
• /
• v.16 no.6
• /
• pp.919-932
• /
• 2015

This paper presents a new method for assessing the three-point-bending (3PB) strength of mortar beams in a non-destructive manner, based on neural network (NN) models. The models are based on the radial basis function (RBF) architecture and the fuzzy means algorithm is employed for training, in order to boost the prediction accuracy. Data for training the models were collected based on a series of experiments, where the cement mortar beams were subjected to various bending mechanical loads and the resulting pressure stimulated currents (PSCs) were recorded. The input variables to the NN models were then calculated by describing the PSC relaxation process through a generalization of Boltzmannn-Gibbs statistical physics, known as non-extensive statistical physics (NESP). The NN predictions were evaluated using k-fold cross-validation and new data that were kept independent from training; it can be seen that the proposed method can successfully form the basis of a non-destructive tool for assessing the bending strength. A comparison with a different NN architecture confirms the superiority of the proposed approach.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.22 no.4
• /
• pp.46-53
• /
• 2008

In this study, we developed RBFN(Radial Basis Function Networks) based load modeling method with harmonic components. The developed method considers harmonic information as well as fundamental frequency and voltage considered as essential factors in conventional method. Thus, the reposed method makes it possible to effectively estimate load characteristics in power lines with harmonics. RBFN has some advantage such as simple structure and rapid computation ability compared with multi-layer perceptorn which is extensively applied for load modeling. To verify the effectiveness, the proposed method has been intensively tested with various dataset acquired under the different frequency and voltage and compared it with conventional methods such as polynomial method, MLPN and RBFN with no harmonic components.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.4A
• /
• pp.318-324
• /
• 2011

In this paper, three coordinated random beamforming (CRBF) schemes are analyzed in a two-cell symmetric interference channel. A simple partial coordination of RBF with base station selection (BSS) is shown to achieve the same average sum rate performance of CRBF with joint encoding (JE). To improve the sum rate performance further, we also propose a transmission mode selection (TMS) between the BSS and JE which is shown to have additional sum rate gain for the large number of users. Simulation results verify the eectiveness of the proposed CRBF schemes and accuracy of the proposed analysis.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.10 no.3
• /
• pp.188-193
• /
• 2000

This paper presents an approach to obtain a reduced neuro-fuzzy model for a plant. The Neuro-Fuzzy Network are compose of the Radial Basis Function Networks with Gausis membership and learned by using temporal back propagation. The dependency in rough set theory is used to eliminate rules. Dependency between the condition membership value of each rule in a model and the output of the plant can allow us to see how much contribution the rule is to identify the plant. While the reduced model maintains the same performance as the original one, the selection algorithm can minimize its complexity and redundancy of the structure.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2001.10a
• /
• pp.535-538
• /
• 2001

Since 1990s, many literatures have shown that connectionist models, such as back propagation, recurrent network, and RBF (Radial Basis Function) outperform the traditional models, MA (Moving Average), AR (Auto Regressive), and ARIMA (Auto Regressive Integrated Moving Average) in time series prediction. Neural based approaches to time series prediction require the enough length of historical measurements to generate the enough number of training patterns. The more training patterns, the better the generalization of MLP is. The researches about the schemes of generating artificial training patterns and adding to the original ones have been progressed and gave me the motivation of developing VTG schemes in 1996. Virtual term is an estimated measurement, X(t+0.5) between X(t) and X(t+1), while the given measurements in the series are called actual terms. VTG (Virtual Tern Generation) is the process of estimating of X(t+0.5), and VTG schemes are the techniques for the estimation of virtual terms. In this paper, the alternative VTG schemes to the VTG schemes proposed in 1996 will be proposed and applied to multivariate time series prediction. The VTG schemes proposed in 1996 are called deterministic VTG schemes, while the alternative ones are called stochastic VTG schemes in this paper.

• International Journal of Computer Science & Network Security
• /
• v.22 no.11
• /
• pp.213-221
• /
• 2022

Intelligent Character Recognition System for Account Payable (ICRS AP) Automation represents the process of capturing text from scanned invoices and extracting the key fields from invoices and storing the captured fields into properly structured document format. ICRS plays a very critical role in invoice data streamlining, we are interested in data like Vendor Name, Purchase Order Number, Due Date, Total Amount, Payee Name, etc. As companies attempt to cut costs and upgrade their processes, accounts payable (A/P) is an example of a paper-intensive procedure. Invoice processing is a possible candidate for digitization. Most of the companies dealing with an enormous number of invoices, these manual invoice matching procedures start to show their limitations. Receiving a paper invoice and matching it to a purchase order (PO) and general ledger (GL) code can be difficult for businesses. Lack of automation leads to more serious company issues such as accruals for financial close, excessive labor costs, and a lack of insight into corporate expenditures. The proposed system offers tighter control on their invoice processing to make a better and more appropriate decision. AP automation solutions provide tighter controls, quicker clearances, smart payments, and real-time access to transactional data, allowing financial managers to make better and wiser decisions for the bottom line of their organizations. An Intelligent Character Recognition System for AP Automation is a process of extricating fields like Vendor Name, Purchase Order Number, Due Date, Total Amount, Payee Name, etc. based on their x-axis and y-axis position coordinates.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.60 no.12
• /
• pp.2352-2360
• /
• 2011

In this paper, we introduce a comprehensive design methodology of Radial Basis Function Neural Networks (RBFNN) that is based on mechanism of clustering and optimization algorithm. We can divide some clusters based on similarity of input dataset by using clustering algorithm. As a result, the number of clusters is equal to the number of nodes in the hidden layer. Moreover, the centers of each cluster are used into the centers of each receptive field in the hidden layer. In this study, we have applied Fuzzy-C Means(FCM) and K-Means(KM) clustering algorithm, respectively and compared between them. The weight connections of model are expanded into the type of polynomial functions such as linear and quadratic. In this reason, the output of model consists of relation between input and output. In order to get the optimal structure and better performance, Particle Swarm Optimization(PSO) is used. We can obtain optimized parameters such as both the number of clusters and the polynomial order of weights connection through structural optimization as well as the widths of receptive fields through parametric optimization. To evaluate the performance of proposed model, NXT equipment offered by National Instrument(NI) is exploited. The situation awareness system-related intelligent model was built up by the experimental dataset of distance information measured between object and diverse sensor such as sound sensor, light sensor, and ultrasonic sensor of NXT equipment.

• Imaging Science in Dentistry
• /
• v.43 no.3
• /
• pp.153-161
• /
• 2013

Purpose: To prevent low bone mineral density (BMD), that is, osteoporosis, in postmenopausal women, it is essential to diagnose osteoporosis more precisely. This study presented an automatic approach utilizing a histogram-based automatic clustering (HAC) algorithm with a support vector machine (SVM) to analyse dental panoramic radiographs (DPRs) and thus improve diagnostic accuracy by identifying postmenopausal women with low BMD or osteoporosis. Materials and Methods: We integrated our newly-proposed histogram-based automatic clustering (HAC) algorithm with our previously-designed computer-aided diagnosis system. The extracted moment-based features (mean, variance, skewness, and kurtosis) of the mandibular cortical width for the radial basis function (RBF) SVM classifier were employed. We also compared the diagnostic efficacy of the SVM model with the back propagation (BP) neural network model. In this study, DPRs and BMD measurements of 100 postmenopausal women patients (aged >50 years), with no previous record of osteoporosis, were randomly selected for inclusion. Results: The accuracy, sensitivity, and specificity of the BMD measurements using our HAC-SVM model to identify women with low BMD were 93.0% (88.0%-98.0%), 95.8% (91.9%-99.7%) and 86.6% (79.9%-93.3%), respectively, at the lumbar spine; and 89.0% (82.9%-95.1%), 96.0% (92.2%-99.8%) and 84.0% (76.8%-91.2%), respectively, at the femoral neck. Conclusion: Our experimental results predict that the proposed HAC-SVM model combination applied on DPRs could be useful to assist dentists in early diagnosis and help to reduce the morbidity and mortality associated with low BMD and osteoporosis.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.39 no.1
• /
• pp.76-84
• /
• 2002

In this paper, it is studied to remove the baseline wander and to minimize the distortion of ST segment in the noise filtering of ECG. In general, the standard filter and adaptive filter are used to remove the baseline wander of the ECG. But the standard filter is limited because the frequency of the baseline signal is variable and the apative filter is difficult to select the reference signal in case of using the adaptive filter. So we proposed a new method of the structure without reference signal using neural networks. To be convinced of the performance of this method, we used ECG data of MIT-BIHs. and obtained the result of the high performance,(-53.3[dB]) than standard filter(-16.3[dB]) and adaptive filter (-44.9[dB]).