• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.8C
• /
• pp.569-574
• /
• 2008

Noise predictive maximum likelihood(NPML) detector embeds noise prediction! whitening process in the branch metric calculation of Viterbi detector and improves the reliability. In this paper, some high-density optical storage channels are examined, and appropriate NPML systems are designed for each channel.

• The Korean Journal of Applied Statistics
• /
• v.23 no.2
• /
• pp.295-304
• /
• 2010

Coating thickness is one of target variables in quality control process in steel industry. To predict coating thickness and to control quality of anti-fingerprint steel coils, ultraviolet-visible spectra are measured. We propose a variable-interval selection procedure based on the variable importance in projection in partial least square model. Using the proposed variable interval selection method, prediction performance gets better in the reduced model than the full model with full spectra absorbance. It is also shown that the first differencing as a data preprocessing technique does work well for the prediction of coating thickness.

• Journal of IKEEE
• /
• v.26 no.3
• /
• pp.494-499
• /
• 2022

It is reported that genome-wide RNA-seq profiles has potential as biomarkers of aging. A number of researches achieved promising prediction performance based on gene expression profiles. We develop an age prediction method based on the transcriptome of human dermal fibroblasts by selecting a proper age interval. The proposed method executes multiple rules in a sequential manner and a rule utilizes a classifier and a regression model to determine whether a given test sample belongs to the target age interval of the rule. If a given test sample satisfies the selection condition of a rule, age is predicted from the associated target age interval. Our method predicts age to a mean absolute error of 5.7 years. Our method outperforms prior best performance of mean absolute error of 7.7 years achieved by an ensemble based prediction method. We observe that it is possible to predict age based on genome-wide RNA-seq profiles but prediction performance is not stable but varying with age.

• The KIPS Transactions:PartD
• /
• v.19D no.1
• /
• pp.21-28
• /
• 2012

Predication of protein interaction sites for monomer structures can reduce the search space for protein docking and has been regarded as very significant for predicting unknown functions of proteins from their interacting proteins whose functions are known. In the other hand, the prediction of interaction sites has been limited in crystallizing weakly interacting complexes which are transient and do not form the complexes stable enough for obtaining experimental structures by crystallization or even NMR for the most important protein-protein interactions. This work reports the calculation of 3D surface patches of complex structures and their properties and a machine learning approach to build a predictive model for the 3D surface patches in interaction and non-interaction sites using support vector machine. To overcome classification problems for class imbalanced data, we employed an under-sampling technique. 9 properties of the patches were calculated from amino acid compositions and secondary structure elements. With 10 fold cross validation, the predictive model built from SVM achieved an accuracy of 92.7% for classification of 3D patches in interaction and non-interaction sites from 147 complexes.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.6
• /
• pp.744-749
• /
• 2016

This paper presents the design of a computer-generated hologram for beam control of an LCOS-based wavelength selective switch, which is the core technology for next-generation ROADM. By introducing a computer-generated hologram instead of general grating patterns to control the LCOS device, we contribute to building a more efficient wavelength selective switch. With the use of phase modulation properties of LCOS devices, we designed the hologram for five-port output and a 40-channel wavelength selective switch. We applied a multi-level phase modulation technique with the Gerchberg-Saxton algorithm to produce the hologram, which is easily scalable to any different type of wavelength selective switch. With an experimental setup, we verified the usability of the hologram designed for five-port output. We also suggest a hologram design technique for beam control of a 40-channel wavelength selective switch.

• The KIPS Transactions:PartD
• /
• v.15D no.6
• /
• pp.785-792
• /
• 2008

In spite of the various techniques on the testing of embedded software, operation failures of embedded systems such as robot or satellite applications, are occurred frequently. The critical reason of these failures is due to the fact that software is embedded into a target system with inherent faults. Therefore, in order to prevent the failure owing to such faults, it needs a technique to test the embedded software which operates in real environment. In this paper, we propose a testing technique, aspect-based On-the-Fly testing that is to test the functionality and performance at real operation time. Our proposed technique gives some benefits of real test of unexpected input conditions, prevention of software malfunction, and reusability of aspect components for the testing.

• Journal of Convergence for Information Technology
• /
• v.11 no.2
• /
• pp.82-89
• /
• 2021

The objective of this study is to propose a new Glide Path that dynamically adjusts the risky asset inclusion ratio of the Target Date Fund by simultaneously considering the market's forecast volatility as well as the time of investor retirement, and to compare the investment performance with the traditional Target Date Fund. Forecasts of market volatility utilize historical volatility, time series model GARCH volatility, and the volatility index VKOSPI. The investment performance of the new dynamic Glide Path, which considers stock market volatility has been shown to be excellent during the analysis period from 2003 to 2020. In all three volatility prediction models, Sharpe Ratio, an investment performance indicator, is improved with higher returns and lower risks than traditional static Glide Path, which considers only retirement date. The empirical results of this study present the potential for the utilization of the suggested Glide Path in the Target Date Fund management industry as well as retirees.

• Proceedings of the Korean Association for Survey Research Conference
• /
• 2001.04a
• /
• pp.87-108
• /
• 2001

Recently, the user of mobile and cordless internet is growing rapidly. So the advertising methodologies are appeared on stage through the internet and mobile service. But the current advertisement(AD) service into mobile phones supplies a short-sentence AD using letter-message, and gives mobile phone users an inconvenience by making them first listen to and then check the AD. Until now the AD service using mobile has been on the beginning level. In this study, we suggest the new advertising methodologies and propose both target marketing strategies and demand forecast through the data mining techniques.

• Survey Research
• /
• v.2 no.1
• /
• pp.87-108
• /
• 2001

Recently, the user of mobile and cordless internet is growing rapidly. So the advertising methodologies are appeared on stage through the internet and mobile service. But the current advertisement(AD) service into mobile phones supplies a short-sentence AD using letter-message, and gives mobile phone users an inconvenience by making them first listen to and then check the AD. Until now the AD service using mobile has been on the beginning level. In this study, we suggest the new advertising methodologies and propose both target marketing strategies and demand forecast through the data mining techniques.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.239-244
• /
• 2023

신경망 기반의 다국어 및 교차 언어 정보 검색 모델은 타겟 언어로 된 학습 데이터가 필요하지만, 이는 고자원 언어에 치중되어있다. 본 논문에서는 이를 해결하기 위해 영어 학습 데이터와 한국어-영어 병렬 말뭉치만을 이용한 효과적인 다국어 정보 검색 모델 학습 방법을 제안한다. 언어 예측 태스크와 경사 반전 계층을 활용하여 인코더가 언어에 구애 받지 않는 벡터 표현을 생성하도록 학습 방법을 고안하였고, 이를 한국어가 포함된 다국어 정보 검색 벤치마크에 대해 실험하였다. 본 실험 결과 제안 방법이 다국어 사전학습 모델과 영어 데이터만을 이용한 베이스라인보다 높은 성능을 보임을 실험적으로 확인하였다. 또한 교차 언어 정보 검색 실험을 통해 현재 검색 모델이 언어 편향성을 가지고 있으며, 성능에 직접적인 영향을 미치는 것을 보였다.