• 정보관리학회지
• /
• 제27권2호
• /
• pp.241-252
• /
• 2010

본 연구에서는 원문헌의 표제가 문헌클러스터링에서 문헌의 주제를 나타내는데 효과적인 자질로 인식되고 있지만 동의어나 유사어를 포함하여 문헌의 주제를 대표하는데 한계가 있음을 인지하고 인용문헌의 표제로 클러스터링 자질을 확대하는 방안을 제시하였다. 문헌 클러스터링의 자질로 원 문헌의 표제 용어와 인용문헌의 표제 용어, 두 종류의 표제 용어를 혼합하여 적용하여 인용문헌의 표제가 클러스터링 성능을 향상시키는 정도를 측정하였다. 각 자질별로 계층적 클러스터링 기법 3개, within group average linkage, complete linkage, Ward 기법을 결합하여 클러스터를 생성하는 성능을 비교, 분석하였는데 원문헌과 인용문헌 표제어를 혼합하여 within group average linkage 기법으로 클러스터링 한 경우가 가장 좋은 결과를 나타내었다.

• 간호행정학회지
• /
• 제13권3호
• /
• pp.302-310
• /
• 2007

Purpose: This study was designed to identify clinical competence and education needs of hospital nurses. Method: A questionnaire was used for data collection which was done form July 1 to July 30, 2005. The respondents were 165 staff nurses in a university hospital. The study instrument were clinical education need scale and Lee(2002)'s clinical competence scale. The data were analyzed using percentages, means, t-test, ANOVA with SPSS-WIN 10.0 program. Result: The results of this study are summarized as follows: The total mean score for clinical competence was 2.57 and education need was 3.30 on a 4 point scale. There was a significant difference in clinical competence according to individual factors of age, marital status, preceptorship experience, present ward, job satisfaction, and total clinical experience. also there was a significant difference in education needs according to preceptorship experience and present ward. Conclusions: Conducting a education needs and clinical competence assessment can provide valuable information that will plan education program and improve staff competence.

• 한국직업건강간호학회지
• /
• 제10권2호
• /
• pp.174-182
• /
• 2001

This study was performed to find out the differences between noise levels of hospital wards and the nurses efforts for noise management in some general hospitals. The hospital wards selected were the intensive care unit(ICU), the emergency room(ER), the nursery room(NR), the internal medicine(IM), the general surgery(GS) among the 5 general hospitals located in Seoul. The data were collected from August 3 to September 13, 1999 through questionnaire survey and noise measurement in each nursing station of hospital wards. Data analysis was done by SPSS 8.0 package among the 305 questionnaires and 24 hours monitored noise levels. Frequency, Chi-square and ANOVA test were used. The study results were as belows: 1. The noise level measured by 24 hours monitoring survey were exceeded on the standard limit in all the hospital wards. Data also showed that noise levels were significantly different in each ward among the three shifts working duties. 2. The subjects were all female nurses. They were mostly working in the ICU ward(28.9%). They were 26~30 years old (43.9%), junior college graduates(57.0%), working for 1~5 years(55.1%) as staff-nurse(85.6%). There were no significant differences between hospital wards and general characteristics of nurses. 3. The noise levels perceived by nurses were regarded as 'Highly noisy'(56.4%), especially during the 11:30 and 15:30 (30.2%) o'clock. Data also showed that noise education was not ever given to nurses(89.9%). Nurses also responded that they hardly put an effort to reduce noise level(54.8%). However, there were significant differences between wards and noisy working time, experience of noise education and level of effort for noise reduction. 4. Nurses also perceived the ventilator alarm and EKG-alarm as the most disturbing sounds in the ICU, human voice and telephone ringing in the ER, human voice and EKG-alarming in the NR, human voices and telephone ringing in IM and GS both wards respectively in order. There were significant differences between hospital wards and noise making factors. 5. Nurses were shown that they regarded highly 'Sound reduction of the human voice', 'Careful handling on medical instruments', and 'Immediate appliances on alarming materials' as the practical method for noise management. There were significant differences between hospital wards and behavioral practical efforts for noise management. According to that results, the statistical differences were shown in the 24 hour monitored noise levels in each ward. Also, nurses perceived the noise severity differently and they approached variously on the practical efforts for noise reduction in each ward. Thus, author thinks that concrete and systematic endeavor will be necessary for noise reduction and management in hospitals for better working and healing environment for both of patients and staffs.

• 한국식품과학회지
• /
• 제23권6호
• /
• pp.711-716
• /
• 1991

조리 식품의 관능특성을 평가하고, 다변수 통계법으로 서로의 상관관계를 조사 하였다. 시료 식품을 각각 특징지을 수 있는 12개의 관능특성을 단계적 차별 분석에 의해 선정하였으며, 요민분석에 의해 도출된 3개의 요인으로 12개의 관능 특성이 갖는 변이의 61.9%를 설명할 수 있었다. 요인 I은 질적 관능성질과 관련이 있고, 요인 II는 양적인 관능특성과 높은 상관관계를 나타내었다. 시료 식품과 관능특성을 동시에 주 성분 좌표상에 표시함으로서 서로간의 상관관계 설정이 용이하였고, average linkage 및 Ward's method을 이용한 집락분석에서 9개의 조리식품은 관능특성의 유사성에서 크게 3개의 집락으로 분류되었다.

• 한국정보과학회논문지:컴퓨팅의 실제 및 레터
• /
• 제15권11호
• /
• pp.856-860
• /
• 2009

본 논문은 Wang-Ward의 WTHE(Weighted and Thresholded Histogram Equalization) 방법에 히스토그램 분할 개념을 적용한 새로운 영상 화질 개선 방법(DWTHE: Decomposable WTHE)을 제안한다. DWTHE는 먼저 영상의 평균 자기 값 또는 명도 균등 분할점을 기준으로 입력 히스토그램의 영역을 분할하고, 분할된 각 영역의 확률밀도 값을 가중치로 사용하여 새로운 히스토그램을 만든 후, 히스토그램 평활화 과정을 수행하게 된다. 하나의 가중치를 사용하는 WTHE 방법과 다르게, 제안 방법은 히스토그램 분할로 인한 복수외 가중치 값을 사용하게 되며, 실험 결과 제안 방법은 기존 방법에 비해 우수한 화질 개선 효과를 보여주었다.

• KEPCO Journal on Electric Power and Energy
• /
• 제2권1호
• /
• pp.61-69
• /
• 2016

Understanding load patterns and customer classification is a basic step in analyzing the behavior of electricity consumers. To achieve that, there have been many researches about clustering customers' daily load data. Nowadays, the deployment of advanced metering infrastructure (AMI) and big-data technologies make it easier to study customers' load data. In this paper, we study load clustering from the view point of yearly and daily load pattern. We compare four clustering methods; K-means clustering, hierarchical clustering (average & Ward's method) and DBSCAN (Density-Based Spatial Clustering of Applications with Noise). We also discuss the relationship between clustering results and Korean Standard Industrial Classification that is one of possible labels for customers' load data. We find that hierarchical clustering with Ward's method is suitable for clustering load data and KSIC can be well characterized by daily load pattern, but not quite well by yearly load pattern.

• International Journal of Computer Science & Network Security
• /
• 제22권7호
• /
• pp.75-84
• /
• 2022

The efficiency of the regional fiscal policy implementation is based on the achievement of target criteria in the formation and distribution of own financial resources of local budgets, reducing their deficit and reducing dependence on transfers. It is also relevant to compare the development of financial autonomy of regions in the course of decentralisation of fiscal relations. The study consists in the cluster analysis of the effectiveness of fiscal policy implementation in the context of 24 regions and the capital city of Kyiv (except for temporarily occupied territories) under conditions of fiscal decentralisation. Clustering of the regions of Ukraine by 18 indicators of fiscal policy implementation efficiency was carried out using Ward's minimum variance method and k-means clustering algorithm. As a result, the regions of Ukraine are grouped into 5 homogeneous clusters. For each cluster measures were developed to increase own revenues and minimize dependence on official transfers to increase the level of financial autonomy of the regions. It has been proved that clustering algorithms are an effective tool in assessing the effectiveness of fiscal policy implementation at the regional level and stimulating further expansion of financial decentralisation of regions.

• Journal of Korean Biological Nursing Science
• /
• 제7권1호
• /
• pp.69-87
• /
• 2005

Purpose : The purpose of this study was to identify pediatric nurses's knowledge related to children's pain and pediatric nurses' career including educational experience and to examine whether there is a difference of their knowledge according to their career. Method : Subjects of this study was 115 pediatric nurses of one university hospital who understood purpose of this study and accepted to participate in this study. Subjects responded questionnaire by self report. The data were gathered for 7 days during the period from October 20 to October '2:7, 2003. The pediatric nurses' knowledge and attitude survey regarding pain developed by Manworren(2001) was used to identify the knowledge related to children's pain of pediatric nurses. Nurses' career consisted of level of education, presently working ward, duration of experience at clinic, pediatric clinic and presently working ward and educational experience related to children's pain during undergraduate courses, ward orientation and continuing education. Educational experience regarding children's pain was measured by whether they had education related to children's pain during undergraduate courses, ward orientation and continuing education or not and their perception on the contents of education was measured by 5 point Likert type scale. SPSS Windows was used to analyze the data. Scores of the knowledge regarding children's pain and career including educational experience of pediatric nurses were presented as mean and standard deviation. Mean percentage of correct answers responded by the subjects was presented under the category of introduction, assessment, pharmacological and nonpharmacological intervention for children's pain. The difference of scores of the knowledge regarding children's pain according to nurses' career was analyzed using t-test, ANOVA and Tukey as a post hoc. Statistical significance was accepted at the level of p<.05. Result : Pediatric nurses had deficient knowledge related to children's pain on the whole and did not trust pain complaint and underestimated the pain. They were short of concrete practical knowledge about pain assessment and pharmacological and nonpharmacological intervention for children's pain. They were excessively concerned with side effect and addiction of analgesic drugs and had a negative view point on nonpharmacological pain intervention. About 50% of the subjects received education regarding children's pain while their undergraduate courses. The number of nurses who received education regarding children's pain while their undergraduate courses was found to be greater among the recently graduated nurses. In ward orientation and continuing education, only less than 10% of the nurses received education regarding children's pain. Nurses perceived their experienced educational content deficient; experienced educational content consisted of pain introduction, assessment and pharmacological and nonpharmacological intervention for children's pain while their undergraduate course, ward orientation and continuing education. There were no significant differences of the knowledge regarding children's pain of pediatric nurses according to level of education, duration of experience at clinic, pediatric clinic and presently working ward. Scores of the knowledge regarding children's pain of nurses at neonatal ward were significantly greater than those of nurses at pediatric intensive care unit. Scores of the knowledge regarding children's pain of pediatric nurses who received education during their undergraduate course were significantly greater than those who did not receive it during their undergraduate courses. Also scores of the knowledge regarding children's pain of pediatric nurses who received one kind of educational experience were significantly greater than those who had no educational experience. Conclusion : Pediatric nurses had deficient knowledge of children's pain and underestimated the pain of children. Also they were short of concrete practical knowledge on pain assessment and intervention for children's pain.

• 한국지역지리학회지
• /
• 제1권1호
• /
• pp.45-60
• /
• 1995

한국의 온대저기압성 강수에 대하여 10년간(1981년$\sim$1990년) 60개 지점의 저기압성 강수와 관련된 43개 강수자료를 이용하여 지역구분을 시도하였다. 먼저 인자분석(因子分析)을 통하여 요인별 고유치를 분석하여 인자점수(因子點數)의 지역분포를 고찰한 후, 군집분석의 Ward법을 적용하여 강수지역을 구분하고 구분지역별로 강수특성을 고찰하였다. 온대저기압성 강수와 관련된 43개 변수를 이용한 인자분석(因子分析)에서 고유치(固有値)가 1.0이상의 요인은 7개이며, 전변동량(全變動量)의 86%를 설명한다. 제1요인은 중서부지역의 강수특성을 설명해주며 기여율이 40.9%로 가장 높게 나타났다. 군집(群集)분석의 Ward법에 의한 한궁의 온대저기압성 강수지역은 (1) 경기 영서북부지역 (2) 영동 울릉도지역 (3) 호서 영서남부지역 (4) 호남 북서제주지역 (5) 남동제주지역 (6) 영남북부지역 (7) 영남남부지역 등 7개의 대지역과, 22개 중지역으로 구분되었다. 더불어 강수지역의 경계선 설정에는 저기압의 이동경로와 유일기류의 방향, 산맥의 주향(走向) 등이 많은 영향을 주고 있음을 알 수 있었다.

• 대한토목학회논문집
• /
• 제29권3D호
• /
• pp.347-356
• /
• 2009

본 연구는 기존의 도로기능분류 정의와 방법론을 벗어나 교통특성에 따른 도로분류 방법론인 도로특성분류를 기초로 분석을 수행하였다. 도로특성분류에 대한 일련의 과정 중에서 다양한 교통특성을 반영하는 설명변수를 기초로 요인점수를 산출하고, 동질한 도로구간을 그룹핑하는 군집화 분석과정과 적정 군집수 도출에 따른 군집결과비교에 본 연구는 초점을 맞추었다. 도로분류를 위해 병합적 계층 군집분석인 Ward법, 비계층적 군집분석인 K-means법, 자율신경 회로망을 이용한 K-SOM을 사용하여 비교분석하였다. 각 군집기법에 대한 결과를 토대로 비교분석한 결과, 군집 수 5 이하에서는 K-means법, 군집 수 14 이상에서는 Kohonen selforganizing maps가 가장 우수한 것으로 나타났으며, 군집수 5~9사이에서는 Ward법과 Kmeans법의 군집 성능이 불규칙한 패턴을 보임에 따라 세밀한 결과분석을 통해 우수성을 결정하는 것이 바람직할 것으로 분석되었다. 본 연구결과는 다양한 교통특성을 고려한 도로구간의 군집 속성을 분석하고 예측하는 분류화 작업에 중요한 기초적인 자료로 사용될 것으로 기대된다.