• Transactions of Materials Processing
• /
• v.27 no.2
• /
• pp.100-106
• /
• 2018

In the continuous casting process, the incidence of transverse surface cracks on the piece may occur by multiple and diverse variables. It is noted that mathematical models may predict only the occurance of the transverse surface cracks, but can require a lot of time (more than three days) to produce a result with this process. This study applied neural networks to predict whether the cracks on the piece surface occurs or does not occur. The computation time was shortened to three minutes, making it applicable to an on-line program, which predicts the non-cracks or cracks of the piece surface in the actual continuous casting process. In addition, the operating conditions to prevent the occurrence of the transverse surface cracks, using decision boundaries were also suggested.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.10 no.4
• /
• pp.49-55
• /
• 2010

In this paper, we propose the symbol decision method considering the non-uniformity in color space and analyze the BER performance of modulation scheme based on color space, comparing with the conventional WDM scheme based on light intensity. Through numerical simulation, we show the BER performance superiority under the condition of AWGN and common mode noise.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.833-834
• /
• 2008

In this paper, a pattern classification method using closest decision method based on the mean of norm in the closet prototype from an input pattern and its k nearest neighbor prototypes is presented to do accurate classification in arbitrary distributed patterns when the number of patterns is very low. Also this method can be used to classify input pattern precisely when the number patterns is very low because this method considers the weight by the difference of variance in prototypes around the discrimination boundary.

• Proceedings of the IEEK Conference
• /
• 1999.11a
• /
• pp.775-779
• /
• 1999

This paper presents design of AGC(Automatic Gain Control) and DC offset remover suitable for cable modem which makes use of QAM(Quadrature Amplitude Modulation) scheme. Since QAM has multi-level signal characteristic, for high-order QAM, the constellation is dense and the distance of decision boundary between adjacent symbols is short. So AGC and DC offset remover must be designed optionally for preventing performance degradation. AGC is designed into feedback type and is related to the STR(Symbol Timing Recovery)and Paff interpolation algorithm. Whereas AGC need to perform average power detection during many symbols by comparison with the reference power, DC offset remover uses only the instant polarity decision such that simple implementation can be achieved with good performance. Though the AGC and DC offset remover are simulated here only for 256 QAM scheme for convenience'sake, it can be applied to other multi-level QAM or PSK modulation scheme.

• Journal of Intelligence and Information Systems
• /
• v.4 no.1
• /
• pp.25-40
• /
• 1998

본 논문에서는 경사진 의사결정 트리(oblique decision tree)에 의해 몇 개의 영역으로 분할된 입력공간(input space)에서 우수한 성능을 발휘할 수 있도록 유전적 프로그래밍 트리들(genetic programming trees)과 연합된 다중 인공신경망 시스템을 개발하였다. 다중 인공신경망인 지역 에이전트들(local agents)은 불할된 영역을 책임지며, 유전적 프로그래밍 트리들로 구성된 경계 에이전트들 (boundary agents)은 불할된 영역의 경계부분만을 담당하게 된다. 본 연방 에이전트 시스템을 이용하여 설계 초기단계의 정보 제한성을 극복하고, 선박 초기설계 단계에서 선박 중앙부 형상설계를 수행하여 범용 설계 시스템으로서의 유용성을 검증하였다.

• Journal of Information Technology Applications and Management
• /
• v.20 no.2
• /
• pp.177-191
• /
• 2013

Social network has been expected to increase the value of social capital through online user interactions which remove geographical boundary. However, online users in social networks face challenges of assessing whether the anonymous user and his/her providing information are reliable or not because of limited experiences with a small number of users. Therefore. it is vital to provide a successful trust model which builds and maintains a web of trust. This study aims to propose a prediction method for the interpersonal trust which measures the level of trust about information provider in Facebook. To develop the prediction method. we first investigated behavioral research for trust in social science and extracted 5 antecedents of trust : lenience, ability, steadiness, intimacy, and similarity. Then we measured the antecedents from the history of interactive behavior and built prediction models using the two decision trees and a computational model. We also applied the proposed method to predict interpersonal trust between Facebook users and evaluated the prediction accuracy. The predicted trust metric has dynamic feature which can be adjusted over time according to the interaction between two users.

• Journal of the Korean Data and Information Science Society
• /
• v.23 no.2
• /
• pp.375-383
• /
• 2012

The purpose of the ensemble methods is to increase the accuracy of prediction through combining many classifiers. According to recent studies, it is proved that random forests and forward stagewise regression have good accuracies in classification problems. However they have great prediction error in separation boundary points because they used decision tree as a base learner. In this study, we use the kernel ridge regression instead of the decision trees in random forests and boosting. The usefulness of our proposed ensemble methods was shown by the simulation results of the prostate cancer and the Boston housing data.

• Proceedings of the KIEE Conference
• /
• 1996.07b
• /
• pp.1301-1303
• /
• 1996

A new automatic frequency control(AFC) tracking algorithm, which we call a rotational decision directed AFC(RDDAFC) is proposed for QPSK demodulation at the digital direct broadcasting satellite(DBS). In order to prevent the presence of the residual phase difference between symbols received at k and k-l by the CPAFC[1] as well as the AFC based on $tan^{-1}$ circuit[2], the RDDAFC rotates the decision boundary for the kth received symbol by the frequency detector output of the (k-1)th received symbol before passing through the cross product discriminator. Test results show that the total pull-in time of the RDDAFC and PLL was 0.13msec under a carrier frequency offset of 2.4MHz when S/N equals 2dB.

• Proceedings of the IEEK Conference
• /
• 1998.10a
• /
• pp.871-874
• /
• 1998

This paper presents a new morphological spatio-temporal segmentation algorithm. The algorithm incorporates intensity and motion information simultaneously, and uses morphological tools such as morphological filters and watershed algorithm. The procedure toward complete segmetnation consists of three steps: joint marker extraction, boundary decision, and motion-based region fusion. By incorporating spatial and temporal information simultaneously, we can obtain visually meaningful segmentation results. Simulation results demonstrates the efficiency of the proposed method.

• Proceedings of the IEEK Conference
• /
• 2003.07e
• /
• pp.2220-2223
• /
• 2003

In this paper we used Support Vector Machines(SVMs) recently proposed as the loaming method, one of Artificial Neural Network, to divide continuous speech into phonemes, an initial, medial, and final sound, and then, performed continuous speech recognition from it. Decision boundary of phoneme is determined by algorithm with maximum frequency in a short interval. Recognition process is performed by Continuous Hidden Markov Model(CHMM), and we compared it with another phoneme divided by eye-measurement. From experiment we confirmed that the method, SVMs, we proposed is more effective in an initial sound than Gaussian Mixture Models(GMMs).