• 전자공학회논문지CI
• /
• 제41권2호
• /
• pp.87-94
• /
• 2004

특징 집합 선택은 학습 알고리즘의 전처리 과정으로 사용되기도 한다. 수집된 자료가 문제와 관련이 없다거나 중복된 정보를 갖고 있는 경우, 이를 학습 모델생성 이전에 제거함으로써 학습의 성능을 향상시킬 수 있다. 또한 탐색 공간을 감소시킬 수 있으며 저장 공간도 줄일 수 있다. 본 논문에서는 특징 집합의 추출과 추출된 특징 집합의 성능 평가를 위하여 엔트로피를 기반으로 한 휴리스틱 함수를 사용하는 새로운 특징 선택 알고리즘을 제안하였다. 탐색 방법으로는 ACS 알고리즘을 이용하였다. 그 결과 학습에 사용될 특징의 차원을 감소시킴으로써 학습 모델의 크기와 불필요한 계산 시간을 감소시킬 수 있었다.

• 한국통신학회논문지
• /
• 제32권11C호
• /
• pp.1066-1072
• /
• 2007

본 논문에서는 동영상 부호화 방식의 움직임 추정을 위한 고속 다중 참조 프레임 선택 기법에 대해 제안한다. 움직임 추정을 위해 다중 프레임을 참조하는 H.264 동영상 부호화 방식의 경우, 블록별로 선택된 최적의 참조 프레임들은 일정한 상관관계를 유지하는 특성이 존재한다. 제안 방식은 이러한 통계적 특성을 이용하여, 임의의 블록 크기의 움직임 추정 모드에서 선택된 최적의 참조 프레임 정보를 이용하여 하위 블록 크기의 움직임 추정 모드에서 사용할 참조 프레임을 예측한다. 실험 결과를 통해 5개의 참조 프레임 사용을 기준으로 제안하는 기법을 사용하였을 때, 움직임 추정에 소요되는 시간은 가용한 모든 참조 프레임에 대해 움직임 추정을 수행하는 방식 대비 평균 60% 감소되었으며, PSNR 및 발생 비트율 측면에서 거의 동일한 성능을 유지함을 확인할 수 있었다.

• 수산해양기술연구
• /
• 제50권3호
• /
• pp.284-291
• /
• 2014

Finely-striate buccinum Buccinum striatissimum, a species of whelks, is caught mainly by pot in the eastern coastal waters of Korea. In order to determine the size selectivity of pot for the species, comparative fishing experiments were conducted near Yeongil Bay from June to September in 2003 using the dome-shaped pots with different five mesh sizes (17.1, 24.8, 35.3, 39.8, and 48.3 mm). The parameters of logistic equation were estimated by the SELECT (Share Each Length's Catch Total) method based on a multinomial distribution. The model with the estimated split parameter was found to fit the catch data best. The master selection curve was estimated to be s (R)=exp (13.044R-16.438)/[1 + exp (13.044R-16.438)], where R is the ratio of shell height to mesh size. The relative shell height of 50% retention was 1.260, and the selection range was 0.168. Enlargement in mesh size of the pot allows more small-sized whelks to escape.

• 한국통신학회논문지
• /
• 제34권10A호
• /
• pp.760-768
• /
• 2009

본 논문에서는 선박에 적용할 e-네비게이션을 위한 HR(high rate)-WPAN(Wireless Personal Area Network) 시스템에 대한 전송 파라미터를 분석하여 최적 전송파라미터 선택방식을 제시한다. 본 논문에서는 IEEE 802.15.3 CSMA/CA기반의 HR-WPAN을 SAN(Ship Area Network)에 적용할 경우에 고려되어야하는 에너지 효율성 측면에서의 전송파라미터들을 분석하고 에너지를 절약할 수 있는 전송파라미터 결정방식에 대하여 각 선박환경별로 분석하여 제안한다. 특히 SAN환경에서 무선채널의 경로손실(path loss)을 결정하는 가장 큰 파라미터인 선박의 다양한 재질을 고려하여 각 HR-WPAN의 전송파라미터별로 에너지 소모량을 분석하여 파라미터 선택방안을 제시한다. 시뮬레이션 결과 선박환경에 따라서 전송률 선택방식, 전송전력 조절 방식 및 데이터 분할크기의 적절한 선택에 따라 에너지 효율성능이 결정됨을 보였다.

• Journal of Biosystems Engineering
• /
• 제34권1호
• /
• pp.8-14
• /
• 2009

When a farmer buys a tractor, the power size of a tractor is determined by various factors such as farm size, farmer's age, farming type, topographical area of farm. Relationships between tractor selection and these factors were found. Three regression models were developed to analyze the relationship. Those models were an OLS-1 model (based on 567 samples having tractors), an OLS-2 model, and a Tobit model (both based on the 1,941 samples). Regression analysis results showed that farm size and farmer's age affected selection of power size for all models at an 1% significance level. It was also shown that some farming types also had significant relationships with the tractor power size. Upland cultivating farmers and livestock farmers had larger tractors than rice cultivating farmers, while orchard farmers had smaller tractors. As for the topographical area, only middle area had significant difference with plain area. Farmers who had a rice-transplanter or a combine had larger tractors than those who didn't.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권3호
• /
• pp.1348-1375
• /
• 2018

Support Vector Machine (SVM) is a well-known machine learning classification algorithm, which has been widely applied to many data mining problems, with good accuracy. However, SVM classification speed decreases with increase in dataset size. Some applications, like video surveillance and intrusion detection, requires a classifier to be trained very quickly, and on large datasets. Hence, this paper introduces two filter-based instance selection techniques for optimizing SVM training speed. Fast classification is often achieved at the expense of classification accuracy, and some applications, such as phishing and spam email classifiers, are very sensitive to slight drop in classification accuracy. Hence, this paper also introduces two wrapper-based instance selection techniques for improving SVM predictive accuracy and training speed. The wrapper and filter based techniques are inspired by Cuckoo Search Algorithm and Bat Algorithm. The proposed techniques are validated on three popular e-fraud types: credit card fraud, spam email and phishing email. In addition, the proposed techniques are validated on 20 other datasets provided by UCI data repository. Moreover, statistical analysis is performed and experimental results reveals that the filter-based and wrapper-based techniques significantly improved SVM classification speed. Also, results reveal that the wrapper-based techniques improved SVM predictive accuracy in most cases.

• Journal of the Korean Statistical Society
• /
• 제22권2호
• /
• pp.325-337
• /
• 1993

This paper propose two test statistics which enable us to proceed the variable selection in Fisher's linear discriminant function for the case of heterogeneous discrimination with equal training sample size. Simultaneous confidence intervals associated with the test are also given. These are exact and approximate results. The latter is based upon an approximation of a linear sum of Wishart distributions with unequal scale matrices. Using simulated sampling experiments, powers of the two tests have been tabulated, and power comparisons have been made between them.

• 한국해양공학회지
• /
• 제17권2호
• /
• pp.77-84
• /
• 2003

The excessive cost of building ships causes the instability of payability, which manage poorly fishing vessels longer than 24m. As a result, an officer evades embarkation and a vicious circle is repeated. In this study, the optimal engine horsepower system for fishing vessels longer than 24m was invented to develop the most efficient engine horsepower, and also a database program for the most efficient engine horsepower has been developed based on the type of and their size.

• 한국경영과학회지
• /
• 제33권2호
• /
• pp.27-44
• /
• 2008

Owl-based ontology is useful to realize the context-aware services which are composed of the distributed and self-configuring modules. Many ontology-based inference engines are developed to infer useful information from ontology. Since these engines show the uniqueness in terms of speed and information richness, it's difficult to ensure stable operation in providing dynamic context-aware services, especially when they should deal with the complex and big-size ontology. To provide a best inference service, the purpose of this paper is to propose a novel methodology of context-aware engine selection in a contextually prompt manner Case-based reasoning is applied to identify the causality between context and inference engined to be selected. Finally, a series of experiments is performed with a novel evaluation methodology to what extent the methodology works better than competitive methods on an actual context-aware service.

• 한국산업보건학회지
• /
• 제31권4호
• /
• pp.521-530
• /
• 2021

Objectives: This was to examine and explore the elements of Size Korea 6th 3D head anthropometric database and to provide basic information for the selection of respirators in Korea. Methods: This was a pilot study for the first year of work in a two-year-project initiated at KOSHA in 2021. 3D head dimensions data were obtained from the Size Korea Center managing the Size Korea 6th 3D national anthropometry survey databases. The 3D head dimensions data, including 45 dimensions, were used in line with ISO standards (e.g., ISO/TS 16976-2) for examinations, comparisons, statistical analyses, etc. Results: A total of 3,088 subjects were finally determined in this study. The main features were: Male subjects were 52.5%; the highest age group was 15-29 at 36.7%; unhealthy weight group based on BMI was 31.7%; and survey area was the capital region. For the 6th 3D head dimensions data with 45 items, the means and standard deviations for 'Face length' were 115.9±7.5 cm for males and 107.3±6.9 cm for females respectively while those for 'Face width' item were not available since there was no such item in the data. Numerous findings were discussed accordingly. Conclusions: This study showed that there were likely requirements for improvements in the 6th 3D head anthropometric data as follows: Standardization of Korean and English terms; addition of head dimensions items missed in the Size Korea survey; and reliability of generalizability for subjects, suggesting that the study results can be used for further studies or improvement of respirator selection in Korea.