• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제10권3호
• /
• pp.79-84
• /
• 2021

고차원의 데이터를 처리하기 위해서는 데이터의 성질을 유지하면서 특징을 잘 반영할 수 있는 특징 추출 방법이 필요하다. 주성분분석 방법은 고차원 데이터에 포함된 정보를 저차원의 데이터로 변환하여 원래 데이터의 변수 수보다 적은 수의 변수로 고차원 데이터를 표현 할 수 있는 방법으로서 데이터의 특징 추출을 위한 대표적인 방법이다. 본 연구에서는 데이터가 고차원인 경우 데이터 특징 추출을 위한 주성분 분석에 있어서 주성분 변수 선정 시 적응적 상관도를 기반으로 한 주성분 분석 방법을 제안한다. 제안하는 방법은 입력 데이터간의 상관 관계를 기반으로 상관도를 적응적으로 반영하여 데이터의 주성분을 분석함으로써 다른 여러 변수에 중복적으로 상관도가 높은 변수와 주성분을 유도하는데 연관성이 적은 변수를 주성분 변수 후보 대상에서 제외시키고자 한다. 고유벡터 계수 값에 의한 주성분 위계를 분석하고 위계가 낮은 주성분이 변수로 선정이 되는 것을 막고 또한 상관 분석을 통하여 데이터의 중복 발생이 데이터 편향을 유도하는 것을 최소화하 하고자 한다. 이를 통하여 주성분 변수 선정 시 데이터 편향성의 영향을 줄임으로써 실제 데이터의 특징을 잘 나타내는 주성분 변수를 선정하는 방법을 제안하고자 한다.

• 보존과학회지
• /
• 제27권2호
• /
• pp.145-154
• /
• 2011

본 연구는 2008년 6월에서 11월까지 국내 9기의 석조문화재를 대상으로 색에 따른 오염등급의 평가와 생물오염피복도를 조사하였다. 생물오염피복도는 홍천물걸리삼층석탑, 수타사삼층석탑 및 수타사홍우당부도에서 90% 정도로 나타났으며, 선태류의 생물오염피복도가 15~23%의 범위로 생물오염이 가장 심한 것으로 나타났다. 색에 따른 생물오염등급은 괘석리사사자삼층석탑, 수타사삼층석탑, 수타사홍우당부도가 등급 5로 오염등급이 가장 높았으며, 여주창리삼층석탑이 등급 2로 양호하게 나타났다. 상관분석 결과 생물오염등급과 지의류의 생물오염피복도는 높은 상관성(r=0.91)을 나타냈지만, 기중조류와 선태류의 생물피복도와는 상관성은 낮게 나타나 이에 대한 보완이 필요할 것으로 판단된다. 각 출현종과 생물오염피복도(선태류, 지의류, 기중조류)의 상관분석결과 선태류의 생물피복도와 Cosmarium decedens, Chlorella ellipsoidea, Anabaena fertilissima, Botryococcus braunii, Chroococcus turgidus, Navicula cryptocephala 및 Xenococcus acervatus가 상관계수 0.56~085의 범위로 높은 상관성을 나타냈다. 지의류의 생물피복도는 Trebouxia sp. 및 Chroococcus varius와 상관계수 0.50 및 0.43으로 나타났으며, 기중조류의 생물오염피복도는 Achnanthes laterostrata 및 Ulothrix zonata와 상관계수 0.76 및 0.57의 상관성을 나타냈다. 생물오염피복도와 상관성이 높은 기중조류는 출현 경향성이 대체로 뚜렷하게 나타나, 지표종으로 유용하게 사용될 수 있을 것이라 판단된다.

• Preventive Nutrition and Food Science
• /
• 제17권1호
• /
• pp.29-35
• /
• 2012

This study investigated the heat level rating of several varieties of Korean red peppers. The chemical constitution of Korean red pepper samples were as follows: 0.54~290.15 mg% capsaicinoids, 79.22~139.09 ASTA value, and 16.76~29.92% free sugar content. The heat level of the Korean red pepper samples was evaluated by trained panelists and the correlation coefficient and F value (0.001%) of the panelist’s results were determined to be significant. In the principle component analysis (PCA), PC1 (capsaicinoids) and PC2 (free sugar) were shown to represent 31.98% and 25.77% of the total variance, respectively. The results of panelists trained for red pepper heat rating were evaluated using analysis of variance and correlation analysis. The trained panelists showed a high F value (p=0.05) and high correlation coefficient. A high correlation efficient of 0.84~0.93 for the test samples with a 40 Scoville heat unit (32,000 SHU red pepper powder) was reported in the sensory evaluation of the Korean red pepper heat level by a trained panel. However, the panel showed a low correlation efficiency of 0.70 $R^2$ when the 60 SHU test samples were included in the analysis.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2008년도 춘계학술대회 논문집
• /
• pp.195-200
• /
• 2008

The accuracy of classifying pixels in HIRIS images is usually degraded by noisy bands since noisy bands may deform the typical shape of spectral reflectance. Proposed in this paper is a statistical method for noisy band removal which mainly makes use of the correlation coefficients between bands. Considering each band as a random variable, the correlation coefficient measures the strength and direction of a linear relationship between two random variables. While the correlation between two signal bands is high, existence of a noisy band will produce a low correlation due to ill-correlativeness and undirectedness. The application of the correlation coefficient as a measure for detecting noisy bands is under a two-pass screening scheme. This method is independent of the prior knowledge of the sensor or the cause resulted in the noise. The classification in this experiment uses the unsupervised k-nearest neighbor algorithm in accordance with the well-accepted Euclidean distance measure and the spectral angle mapper measure. This paper also proposes a hierarchical combination of these measures for spectral matching. Finally, a separability assessment based on the between-class and within-class scatter matrices is followed to evaluate the performance.

• 대한원격탐사학회지
• /
• 제25권3호
• /
• pp.263-270
• /
• 2009

Noise band removal is a crucial step before spectral matching since the noise bands can distort the typical shape of spectral reflectance, leading to degradation on the matching results. This paper proposes a statistical noise band removal method for hyperspectral data using the correlation coefficient between two bands. The correlation coefficient measures the strength and direction of a linear relationship between two random variables. Considering each band of the hyperspectral data as a random variable, the correlation between two signal bands is high; existence of a noisy band will produce a low correlation due to ill-correlativeness and undirected ness. The unsupervised k-nearest neighbor clustering method is implemented in accordance with three well-accepted spectral matching measures, namely ED, SAM and SID in order to evaluate the validation of the proposed method. This paper also proposes a hierarchical scheme of combining those measures. Finally, a separability assessment based on the between-class and the within-class scatter matrices is followed to evaluate the applicability of the proposed noise band removal method. Also, the paper brings out a comparison for spectral matching measures. The experimental results conducted on a 228-band hyperspectral data show that while the SAM measure is rather resistant, the performance of SID measure is more sensitive to noise.

• 한국전문물리치료학회지
• /
• 제15권2호
• /
• pp.20-29
• /
• 2008

The purpose of this study was to examine the validity and reliability of the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC)-VA3.0 in patients with hip and knee osteoarthritis (OA). The sample consisted of 301 patients who had received treatments at the physical therapy units of 5 medical institutions in Andong City in june 2006. Questionnaires on the WOMAC were recruited by 12 physical therapists. The internal structure and reliability of the scales were evaluated by means of item-internal consistency (Cronbach's alpha coefficient: ${\alpha}$), item-discriminant validity, and Pearson's relation coefficient. To explore construct validity, we conducted a principal component factor analysis with varimax rotation analysis. The criterion for factor extraction was an eigenvalue >1.0. The average age of the patients was 62.1 years. All WOMAC subscales (pain, stiffness, and physical function) were internally consistent with Cronbach's coefficients of .81, .91, and .80, respectively. The internal consistency reliability of item-each scale were also internally consistent with Cronbach's coefficient of .89 (Pearson's correlation coefficient: .71~.84), .93 (.89~.91), and .96 (.67~.91), respectively. However, high correlation was found among 3 items (.66~.83, .66~.67, and .67~.83), so the item-discriminant validity was low (${\alpha}$ coefficient: .81, .91, .80, respectively). The construct validity by factor analysis was low because it was not consistent With WOMAC-VA3.0. In conclusion, the results reported here confirm the reliability of the WOMAC in patients with OA of the hip and knee. The collection of information on the hip and knee osteoarthritis using this instrument was acceptable to patients. A further prospective multi-center study will be necessary to prove the construct validity.

• 대한기계학회논문집B
• /
• 제26권12호
• /
• pp.1681-1691
• /
• 2002

The heat transfer and pressure drop characteristics of heat exchangers with louver fins were experimentally investigated. The samples had small fin pitches (1.0 mm to 1.4 mm), and experiments were conducted up to a very low frontal air velocity (as low as 0.3 m/s). At a certain Reynolds number (critical Reynolds number), the flattening of the heat transfer coefficient curve was observed. The critical Reynolds number was insensitive to the louver angle, and decreased as the louver pitch to fin pitch ratio (L$_{p}$F$_{p}$) decreased. Existing correlations on the critical Reynolds number did not adequately predict the data. It is suggested that, for proper assessment of the heat transfer behavior, the louver pattern in addition to the flow characterization need to be considered. The heat transfer coefficient increased as the fin pitch decreased. At low Reynolds numbers, however, the trend was reversed. Possible explanation is provided considering the louver pattern between neighboring fins. Different from the heat transfer coefficient, the friction factor did not show the flattening characteristic. The reason may be attributed to the form drag by louvers, which offsets the decreased skin friction at a low Reynolds number. The friction factor increased as the fin pitch decreased and the louver angle increased. A new correlation predicted 92% of the heat transfer coefficient and 90% of the friction factor within $\pm$10%.10%.

• 한국축산식품학회지
• /
• 제21권2호
• /
• pp.163-168
• /
• 2001

This study was conducted to analyze the errors occurred between standard plate count(SPC) and Bactometer, Malthus, Bactoscan and to investigate correlation coefficient(r) between SPC and each equipments for the purpose of making new calibration curve. Correlation coefficients of three different types of equipments to SPC value were 0.71, 0.81, 0.84 respectively(n=287) and that of three types of equipments were 0.71∼0.82 relatively low. In raw milk, correlation coefficients of three types of equipments of SPC were in the range of 0.67∼0.73 below 3.0${\times}10^4$(CUF/ml) of bacterial number, and those between the three types of equipments were 0.68∼0.72. Between 3.0${\times}10^4$∼5.0${\times}10^5$(CFU/ml) of bacterial number, correlation coefficients of the three different types of equipments were 0.71∼0.81. and those between the three different types of equipments were 0.64∼0.77. Over 5.0${\times}10^5$(CFU/ml), correlation coefficient of the three types equipments were 0.66∼0.83 and those between them were 0.70∼0.85 respectively. The error of the three different types equipments to SPC value was significantly high, 37∼53% of them was under 50% of error range, and in case of raw milk less than 3.0${\times}10^4$ of bacterial number, 44∼67% of them showed over 100% of error.

• 대한물리의학회지
• /
• 제9권3호
• /
• pp.279-284
• /
• 2014

PURPOSE: This study intended to verify whether there was actual correlation between weight-bearing asymmetry and a limitation in hip joint rotation range in patients with low back pain. METHODS: Thirty five low back pain patients voluntarily participated this study. For each participant, hip joint medial rotation symmetry rate and the weight-bearing symmetry rate were calculated. The correlation between the two variables was investigated. RESULTS: A decrease in the left hip joint medial rotation range of motion (ROM) was observed more often than a reduction in the right hip joint medial rotation ROM. However, similar number between right and left side was observed in ground reaction force more weighted. The coefficient between the passive hip joint medial rotation symmetry rate and the weight loading symmetry ratio was -0.19 (p < 0.05). CONCLUSION: The present study demonstrated a weak correlation between the hip joint medial rotation ROM and the weight distribution of both feet. Such result suggests that careful evaluation by separating each element is needed in treating patients with low back pain. Future research should take into account asymmetric alignment and abnormal movement in different joints of the body as well as asymmetry in the bilateral hip joint rotation and the unilateral weight supporting posture.

• Journal of the Korean Data and Information Science Society
• /
• 제22권1호
• /
• pp.57-64
• /
• 2011

이 연구에서는 최근 체육영재선정에 공통적으로 사용되고 있는 스포츠영재성 검사 항목과 코스타스 점수간의 연관성 분석을 통해 상대적으로 중요한 검사 항목과 그렇지 않은 검사 항목을 통계적으로 밝혀내는 데 목적이 있다. 이 연구를 위해 2010년도 서울지역 (강동구, 송파구, 노원구, 광진구 등)과 경기지역 (의정부시, 가평군, 구리시 등) H대학 체육영재양성사업에 지원한 201명의 초등학교 학생들의 자료가 사용되었으며, 육상 네 종목 (투척, 도약, 단거리, 중장거리), 수영 및 체조 종목의 코스타스 점수와 1차 스포츠영재성 검사에서 얻은 기본 체격, 체력 운동능력, 생리적 기능 정보를 이용하였다. 구체적으로 순수한 연관성 분석을 위해 피어슨상관계수와 편상관계수를 사용하였으며, 중요한 검사 항목 선택을 위해 단계적 회귀분석을 이용하였다. 그 결과 골연령, 체중, 흉위와 50m달리기, 하프스쿼트점프가 6개 코스타스 점수에 미치는 영향이 상대적으로 미약한 것으로 나타났다.