• 한국지반공학회지:지반
• /
• 제14권1호
• /
• pp.29-36
• /
• 1998

본 논문에서는 이분적 암반분류 방법 보다 일반적인 다분적 암반분류 방법의 응용에 관해 연구하였다. 특히, 정성적 데이타를 체계적으로 이용할 수 있는 방법이 모색되었다. 응용 예를 통해 Bieniawski의 암반평가 시스템 (rock mass rating system, RMR)과 같이 암반을 두개 이상의 다등급으로 분류할 경우 본 논문에 제시된 방법이 효과적으로 사용될 수 있고 체계적인 암반조사를 위해 크게 기여할 것으로 생각된다. 또한, 오차에 대응하는 비용(cost of errors)의 기대값이 암반조사를 위한 시추 방법이 잘 계획되었는지에 관한 평가척도로 이용될 수 있음을 알았다.

• 대한치과교정학회지
• /
• 제22권1호
• /
• pp.17-30
• /
• 1992

Since 1984, many patients have been treated with Multiloop Edgewise Archwire (MEAW) Technique and diagnosed with ODI (Overbite Depth Indicator) and APDI (Anteroposterior Dysplasia Indicator) by the authors. 234 samples of them were selected randomly for the statistical analysis (age, sex, Angle's classification, treatment period, extraction, ODI etc.). Especially, ODI was analysed statistically and its application methods were reviewed. The results and conclusions were as follows: 1. On the 150 patients with normal overbite, the mean values of Class I, II, III malocclusion were $67.5^{\circ}$, $72.2^{\circ}$ and $59.0^{\circ}$. They were significantly different on the level of p < 0.01. 2. In normal overbite samples, ODI decreased with the increase of APDI and the correlation coefficient was -0.54. It seems that this result reflects the characteristics of AB to mandibular plane angle. 3. The regression equation was Y = - 0.57X + 114.64, where X is APDI and Y is ODI. In cases of small or large APDI, it seems to be absurd that the patient's ODI is compared with the mean ODI to differentiate diagnostically the open bite or deep bite tendency from the normal.

• 한국의료질향상학회지
• /
• 제22권1호
• /
• pp.29-37
• /
• 2016

Readmission which reflects capacity to manage patients and general level of medical services has been known for one of the causes of medical expenditure due to inefficient service. Compared to disease-specific readmission, hospital wide readmission (HWR) is relatively easy to understand, and has merit to get over limitation of collateral medical services assessment; therefore, a growing interest in development and usage of readmission indicator as quality of care indicator focusing on all-disease is detected. In this study, we investigate current state of risk standardized readmission rate indicator used in the United States, the United Kingdom, and Canada, and examine the considerations when using readmission rate as quality indicator in Korea. Differences in risk-adjustment methods were showed among countries. The United States do not control race not to hide socio-demographic factors on readmission. Canada shows differentiation compared to other countries about reflecting community factors. All three-countries utilize readmission rate as monitoring quality of care rather than incentives or penalty due to the fact that readmission rate could not represent the whole quality of hospital and has a limitation at controlling socio-economic factors. Therefore, for usage readmission rate as quality indicator in Korea, preparing readmission classification standard for Korean medical environment and additional methods for acquiring information by using discharge summary is need. Moreover, continued discussion with clinical specialists is needed for obtain clinical reliability and validity.

• 간호행정학회지
• /
• 제6권2호
• /
• pp.245-257
• /
• 2000

The purpose of this study was to validate caregiver outcomes included Nursing Outcomes Classification(NOC) developed by Johnson and Maas at the University of Iowa. A sample of 73 nurse experts working in university affiliated hospitals participated in this study. They were asked to rate indicators that examplified the outcomes on a scale of 1(indicator is not at all characteristic) to 5(indicator is very characteristic). A questionnaire with an adaptation of Fehring's methodology was used to establish the content validity of outcomes. The results were as follow: 1. Eight outcome label were considered to be 'supporting' and three outcome label were considered to be 'nonsupporting'. 2. 'Caregiver-Patient Relationship' attained an OCV score of 0.64 and the highest OCV score among caregiver outcomes.. 3. 'Caregiver Emotional Health' attained an OCV score of 0.54 and the lowest OCV score among caregiver outcomes. Replication study will be needed and outcomes sensitive to Korean culture need to be developed.

• 자원환경지질
• /
• 제40권3호
• /
• pp.319-330
• /
• 2007

청송 무포산응회암은 경상분지 북동부 화산암류에서 하나의 냉각단위로 구별되는 층서단위이다. 이 응회암은 그 입도에 의하면 대부분 응회암에 속하고 구성원에 의하면 파리질 응회암에 속한다. 대부분 부석과 샤드가 일정하게 배열되고 심하게 편평화되어 용결엽리를 발달시키며 여러 가지 흐름지시자를 보여준다. 무포산응회암에서 부석 정향배열에 의한 유상선구조로부터 이동 패턴, 암편과 부석의 와상배열, 용결엽리의 비대칭 유상습곡 등의 흐름 지시자, 암편과 부석의 최대입경의 측방점이, 그리고 구성원 함량의 측방점이에 의하면 이 응회암을 집적시킨 화쇄류는 남동부에서 공급되었던 것으로 해석된다.

• 교육시설 논문지
• /
• 제23권4호
• /
• pp.3-10
• /
• 2016

This paper presents the investigation results on the classification of the university campuses. For the classification, we selected the spatial index as the evaluation indicator since the environmental factors and maintenance methods vary from university campus to university campus. For the study, we used eight spatial indices of the 30 national universities. This paper provides the spatial characteristics of different campus types, presents campus classification analysis as a future research approach to campus maintenance, and provides the data for the future study of comparison among universities. The results are as follows. 1) The classification investigation categorized the university campuses into three groups. Type 1 is a large-scale type, located near downtown. Type 2 is a medium-scale type, located at a remote site from downtown. Type 3 is a small-scale type, which is located comparatively near downtown. 2) Type 1 is a large-scale mixed area type, and 13 universities belong to this group. Type 2 is a medium-scale suburban area type, and six universities are in this group. Finally, Type 3 is a small-scale downtown area type, and 11 universities belong to this group.

• 터널과지하공간
• /
• 제19권5호
• /
• pp.407-420
• /
• 2009

이 연구에서는 시추 조사와 물리탐사 자료와 같은 다양한 지반 정보를 통합하여 최적의 암반 분류 기법을 도출하는데 지시크리깅을 적용하였다. 최적의 지시크리깅 결과를 얻기 위해서는 효과적으로 hard data(시추조사)와 soft data(물리탐사 자료)를 통합하기 위한 알맞은 방법을 모색할 필요가 있다. 이론적인 베리오그램 모델변수를 결정하기 위해 반복적 비선형 역산 방법을 적용하였고 이 알고리즘의 타당성 검증을 위해 목적함수의 분포양상을 검토한 결과 상관거리에 따른 구배는 대단히 작은 특성을 보였다. 현장 적용지역은 지표에서 터널 계획고까지의 심도가 최대 500 m인 대규모 산악터널 예정지이다. 지시크리깅을 이용하여 soft data인 AMT (Audio frequency Magneto-Telluric) 탐사 자료와 hard data인 RMR자료를 하나로 통합하고자 하였다. 결론적으로 터널계획고와 터널 상부 1D 구간에 대한 암반등급도를 작성하여 도시하였다.

• 한국멀티미디어학회논문지
• /
• 제21권5호
• /
• pp.617-625
• /
• 2018

Phishing website has become a crucial concern in cyber security applications. It is performed by fraudulently deceiving users with the aim of obtaining their sensitive information such as bank account information, credit card, username, and password. The threat has led to huge losses to online retailers, e-business platform, financial institutions, and to name but a few. One way to build anti-phishing detection mechanism is to construct classification algorithm based on machine learning techniques. The objective of this paper is to compare different classifier ensemble approaches, i.e. random forest, rotation forest, gradient boosted machine, and extreme gradient boosting against single classifiers, i.e. decision tree, classification and regression tree, and credal decision tree in the case of website phishing. Area under ROC curve (AUC) is employed as a performance metric, whilst statistical tests are used as baseline indicator of significance evaluation among classifiers. The paper contributes the existing literature on making a benchmark of classifier ensembles for web phishing detection.

• Journal of Multimedia Information System
• /
• 제5권2호
• /
• pp.99-104
• /
• 2018

Phishing website has become a crucial concern in cyber security applications. It is performed by fraudulently deceiving users with the aim of obtaining their sensitive information such as bank account information, credit card, username, and password. The threat has led to huge losses to online retailers, e-business platform, financial institutions, and to name but a few. One way to build anti-phishing detection mechanism is to construct classification algorithm based on machine learning techniques. The objective of this paper is to compare different classifier ensemble approaches, i.e. random forest, rotation forest, gradient boosted machine, and extreme gradient boosting against single classifiers, i.e. decision tree, classification and regression tree, and credal decision tree in the case of website phishing. Area under ROC curve (AUC) is employed as a performance metric, whilst statistical tests are used as baseline indicator of significance evaluation among classifiers. The paper contributes the existing literature on making a benchmark of classifier ensembles for web phishing detection.

• 한국지구과학회지
• /
• 제31권4호
• /
• pp.301-312
• /
• 2010

지화학 자료는 환경 관리를 위한 중요한 환경 변수중 하나로 인식되어 왔다. 지화학 자료는 보통 공간적으로 산재되어 수집되기 때문에, 샘플링 되지 않은 지점에서의 속성값 예측과 더불어 부가적인 분석을 위해 예측에 수반되는 불확실성을 추정할 필요가 있다. 이 논문은 지시자 지구통계학이 지화학 자료의 공간적인 분포값의 제시뿐만 아니라 의사결정을 보조할 수 있는 정보를 제공하기 위해 유용하게 사용될 수 있는지를 예시하고자 한다. 카드뮴 자료의 추정사례 연구를 통해 확률론적 불확실성 모델링, 위험성 분석 등 지구통계학적 분석의 틀을 제시하였다. 지시자 크리깅을 통해 조건부 누적 분포 함수를 모델링한 후에, 기대값 추정치와 조건부 분산을 카드뮴의 추정값과 정량적 불확실성 추정을 위해 각각 계산하였다. 그리고 확률 임계치와 속성 임계치의 적용을 통해 오염/비오염 지역을 구분하였다. 또한 조건부 분산과 속성값과 임계치값의 차이를 모두 설명할 수 있는 변동 계수를 통해 추가적인 샘플링 지점을 추출하였다. 이 연구에서 적용한 지시자 지구통계학적 분석 틀은 불확실성을 고려한 의사 결정과 관련하여 지화학 자료를 포함한 환경 변수의 분석에 유용하게 사용될 수 있을 것으로 기대된다.