• The KIPS Transactions:PartD
• /
• v.11D no.3
• /
• pp.709-716
• /
• 2004

This paper describes an intelligent real-time monitoring system of a semiconductor processing equipment, which determines normal or not for a wafer in processing, using multiple time-series pattern recognition. The proposed system consists of three phases, initialization, learning and real-time prediction. The initialization phase sets the weights and tile effective steps for all parameters of a monitoring equipment. The learning phase clusters time series patterns, which are producted and fathered for processing wafers by the equipment, using LBG algorithm. Each pattern has an ACI which is measured by a tester at the end of a process The real-time prediction phase corresponds a time series entered by real-time with the clustered patterns using Dynamic Time Warping, and finds the best matched pattern. Then it calculates a predicted ACI from a combination of the ACI, the difference and the weights. Finally it determines Spec in or out for the wafer. The proposed system is tested on the data acquired from etching device. The results show that the error between the estimated ACI and the actual measurement ACI is remarkably reduced according to the number of learning increases.

• Food Engineering Progress
• /
• v.21 no.1
• /
• pp.72-78
• /
• 2017

The objective of this study was to investigate the effects of combinations of tofu paste and non-starch polysaccharides (NSP) on the oil uptake reduction (OTR) of deep-fat fried cake doughnuts. OTR agents were tofu paste (from grinding tofu with deionized water, followed by passage through a 60 mesh sieve), and five neutral and nine anionic NSPs. A control doughnut (without tofu paste or NSP), tofu doughnut (with tofu paste) and NSP-tofu doughnut (with tofu paste and NSP) were prepared. The moisture and total lipid (TL) content, cross-section image, color characteristic, and specific volume were measured. The tofu and NSP-tofu doughnuts exhibited higher moisture and lower TL content than the control. OTR was 10.8% for the tofu doughnut, and between 13.2% and 41.2% for the NSP-tofu doughnut. The highest OTR (41.2%) was found in the NSP-tofu doughnut with a combination of tofu paste and sodium alginate (NaA). The specific volume of the NSP-tofu doughnuts with combinations of tofu paste with NaA (2.5 mL/g), locust bean gum (2.5 mL/g), and ${\kappa}$-carrageenan (2.4 mL/g) was very close to that of the control (2.6 mL/g). Considering the OTR and specific volume of doughnuts, the combination of tofu paste and NaA would be most effective in reducing the oil uptake of doughnuts during deep-fat frying.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.06c
• /
• pp.95-98
• /
• 1998

본 논문에서는 이산적인 HMM(Hidden Markov Model)을 이용한 고립단어 인식 시스템에서 입력특징 벡터의 변별력을 향상시키기 위해 수정된 집단화 알고리듬을 제안하므로써 K-means나 LBG 알고리듬을 이용한 기존의 HMM에 비해 2.16%의 인식율을 향상시켰다. 또한 HMM학습과정에서 불충분한 학습데이타로 인해 발생되는 인식율저하의 문제를 해소하기 위해 개선된 smoothing 기법을 제안하므로써 화자독립 실험에서 3.07%의 인식율을 향상시켰다. 본 논문에서 제안한 두가지 알고리듬을 모두 적용하여 최종적으로 실험한 VQ/HMM에서는 기존의 방식에 비해 화자독립 인식실험 결과 평균 인식율이 4.66% 개선되었다.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.26 no.11
• /
• pp.1857-1866
• /
• 1989

In this paper, an algorithm to update a codebook using a neural network in consecutive images, is proposed. With the Kohonen's self-organizing feature map, we adopt the iterative technique to update a centroid of each cluster instead of the unsupervised learning technique. Because the performance of this neural model is comparable to that of the LBG algorithm, it is possible to update the codebooks of consecutive frames sequentially in TV and to realize the hardwadre on the real-time implementation basis.

• Journal of Life Science
• /
• v.14 no.3
• /
• pp.391-395
• /
• 2004

To overproduce the cyclodextrin glucanotransferase(CGTase) of Bacillus stearothermophilus NO2 in B. subtilis, the pJH-CGTl plasmid (8.14 kb) was constructed, in which the ORF of CGTase gene could be transcribed by strong constitutive promoter, P$\_$JH/. To overproduce CGTase from a recombinant B. subtilis, the effect of media on the cell growth and expression level of CGTase were investigated with various media (LB, 2${\times}$LB, 5% molasses+2% CSL, CS, LBG) in the flask culture. Among them, [5% molasses+2% CSL] medium revealed the maximum expression level of CGTase with 1.8 unit/$m\ell$ at 9 hr culture. In the batch culture on [10% molasses+5% corn steep liquor] medium the expression level of CGTase, the secretion efficiency, and plasmid stability were about 4.2 unit/$m\ell$, 90% and 90%, respectively, at 30 hr culture. The cell growth and expression level in the fermenter culture with the industrial molasses medium were increased by 2-folds over the flask culture.

• Journal of Internet Computing and Services
• /
• v.6 no.1
• /
• pp.105-114
• /
• 2005

This paper describes a new measuring technique by analysing multiple time-series patterns. This paper's goal is that extracts a really measured value having a sample pattern which is the best matched with an inputted time-series, and calculates a difference ratio with the value. Therefore, the proposed technique is not a recognition but a measurement. and not a hardware but a software. The proposed technique is consisted of three stages, initialization, learning and measurement. In the initialization stage, it decides weights of all parameters using importance given by an operator. In the learning stage, it classifies sample patterns using LBG and DTW algorithm, and then creates code sequences for all the patterns. In the measurement stage, it creates a code sequence for an inputted time-series pattern, finds samples having the same code sequence by hashing, and then selects the best matched sample. Finally it outputs the really measured value with the sample and the difference ratio. For the purpose of performance evaluation, we tested on multiple time-series patterns obtained from etching machine which is a semiconductor manufacturing.

• The Journal of Industrial Distribution & Business
• /
• v.11 no.4
• /
• pp.31-38
• /
• 2020

Purpose: At a time when the distribution industry is dominated by capital and technology, win-win growth among businesses groups (BGs) in wholesale market is becoming a social issue. Therefore, through analysis of market growth, market concentration (MC) and market power (MP), we want to identify the structure of the wholesale market and the competitiveness of the BGs in terms of market share (MS), sales-profit ratio (SPR), and labor productivity (LP) to explore the possibility of win-win growth. Market situation: Wholesale and Retail sales ratio (W/S) continues to increase, which also means inefficiency in distribution channels or opportunities in wholesale markets. Wholesale sales have grown 8.3 percent annually over the past 15 years, while the number of companies and workers has declined since 2017, which is why some restructuring is believed to begin in the wholesale industry. In terms of MC and MP, the growth potential of SBG can be found in FCB, ARM, FBT and CME BTs. Methodology and data: Through ANOVA and Regression Analysis, the 2015 Economic Census Data of KOSTAT was analyzed. Results: The results of ANOVA show that statistically significant SBG has a larger MS than LBG. The SPR was not different among BGs. LP is higher for LBG than for other BGs. Regression results show that the employment weight (EW) and the company size (SC) have positive effects on the MS, but the company weight (CW) and employment size (SE) have negative effects. In the case of SPR, the CW is positive and the EW is negative. In addition, LP appears to be more positive as SC in the BGs is larger. Conclusions: Although there is sufficient potential for SBG in the wholesale market, there is a problem that needs to increase LP. Therefore, the SBG needs to restructure in terms of number of companies and SC to improve the efficiency of employment. In terms of MC and MP, the SBG looks for possibilities in FCB, ARM, FBT and CME BTs. In addition, SBG that seeks higher returns with human services rather than simple sales is found to be competitive in the HHG, MES and CME BTs.

• Journal of Communications and Networks
• /
• v.5 no.2
• /
• pp.157-166
• /
• 2003

This paper deals with the problem of channel identification for Single Input Multiple Output (SIMO) slow fading channels using clustering algorithms. Due to the intrinsic memory of the discrete-time model of the channel, over short observation periods, the received data vectors of the SIMO model are spread in clusters because of the AWGN noise. Each cluster is practically centered around the ideal channel output labels without noise and the noisy received vectors are distributed according to a multivariate Gaussian distribution. Starting from the Markov SIMO channel model, simultaneous maximum ikelihood estimation of the input vector and the channel coefficients reduce to one of obtaining the values of this pair that minimizes the sum of the Euclidean norms between the received and the estimated output vectors. Viterbi algorithm can be used for this purpose provided the trellis diagram of the Markov model can be labeled with the noiseless channel outputs. The problem of identification of the ideal channel outputs, which is the focus of this paper, is then equivalent to designing a Vector Quantizer (VQ) from a training set corresponding to the observed noisy channel outputs. The Linde-Buzo-Gray (LBG)-type clustering algorithms [1] could be used to obtain the noiseless channel output labels from the noisy received vectors. One problem with the use of such algorithms for blind time-varying channel identification is the codebook initialization. This paper looks at two critical issues with regards to the use of VQ for channel identification. The first has to deal with the applicability of this technique in general; we present theoretical results for the conditions under which the technique may be applicable. The second aims at overcoming the codebook initialization problem by proposing a novel approach which attempts to make the first phase of the channel estimation faster than the classical codebook initialization methods. Sample simulation results are provided confirming the effectiveness of the proposed initialization technique.

• The Journal of the Acoustical Society of Korea
• /
• v.11 no.1
• /
• pp.22-34
• /
• 1992

There exists a transform trellis code that is optimal for stationary Gaussian sources and the squared-error distortion measure at all rates. In this paper, we train an asymptotically optimal version of such a code to obtain one which is matched better to the statistics of real world data. The training algorithm uses the M algorithm to search the trellis codebook and the LBG algorithm to update the trellis codebook. We investigate the trained transform trellis coding scheme for the first-order AR(autoregressive) Gaussian source whose correlation coefficient is 0.9 and actual speech sentences. For the first-order AR source, the achieved SNR for the test sequence is from 0.6 to 1.4 dB less than the maximum achievable SNR as given by Shannon's rate-distortion function for this source, depending on the rate and surpasses all previous known results for this source. For actual speech data, to achieve improved performance, we use window functions and gain adaptation at rate 1.0 bits/sample.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.6 no.1
• /
• pp.34-42
• /
• 2011

In an effort to the practical realization of phonetic typewriter, we concentrate on the classification of consonants in this paper. Since many of consonants do not show periodic behavior in time domain and thus the validity for Fourier analysis of them are not convincing, vector quantization (VQ) via LBG clustering is first performed to check if the feature vectors of MFCC and LPCC are ever meaningful for consonants. Experimental results of VQ showed that it's not easy to draw a clear-cut conclusion as to the validity of Fourier analysis for consonants. For classification purpose, two kinds of neural networks are employed in our study: self organizing map (SOM) and learning vector quantization (LVQ). Results from SOM revealed that some pairs of phonemes are not resolved. Though LVQ is free from this difficulty inherently, the classification accuracy was found to be low. This suggests that, as long as consonant classification by LVQ is concerned, other types of feature vectors than MFCC should be deployed in parallel. However, the combination of MFCC/LVQ was not found to be inferior to the classification of phonemes by language-moded based approach. In all of our work, LPCC worked worse than MFCC.