• Communications for Statistical Applications and Methods
• /
• 제20권3호
• /
• pp.225-233
• /
• 2013

The linear support vector machine(SVM) is motivated by the maximal margin separating hyperplane and is a popular tool for binary classification tasks. Many studies exist on the consistency properties of SVM; however, it is unknown whether the linear SVM is consistent for estimating the optimal classification boundary even in the simple case of two Gaussian classes with a common covariance, where the optimal classification boundary is linear. In this paper we show that the linear SVM can be inconsistent in the univariate Gaussian classification problem with a common variance, even when the best tuning parameter is used.

• 사상체질의학회지
• /
• 제18권2호
• /
• pp.96-112
• /
• 2006

1. Objectives This study focuses on the reliability of the Questionnaire about Sasang Constitution Classification for Mongolians 2. Methods Test-retest method and internal consistency method have been performed based on the absolutely diagnosed group of 87 cases who respond to the questionaries during the time interval of one yeat between July 2003 and July 2004 to verify the confidence level. 3. Results and Conclusions (1) In the test-retest for each question of the Questionnaire of Sasang Constitution Classification for Mongolians, the dependency ratio is 40% and the agreement ratio is 92%. Therefore, this questionnaire has credibility because it has question relations and high agreement ratio. (2) In the internal consistency method for the measure of the Questionnaire of Sasang Constitution Classification for Mongolians, the value of Cronbach alpha is mote than 0.60. As a result, this questionnaire has internal consistency for each question which explains each physical constitution and it has credibility (3) In the internal consistency method for the measure of the Questionnaire of Sasang Constitution Classification for Mongolians, the Pearson's correlation coefficient, r, falls between $+0.38\;{\sim}\;+0.54$ in each measure. Accordingly, this questionnaire has internal consistency between each physical constitution measure and it has credibility.

• Journal of Gastric Cancer
• /
• 제11권2호
• /
• pp.78-85
• /
• 2011

Since January of 2010, the seventh edition of UICC tumor node metastasis (TNM) Classification, which has recently been revised, has been applied to almost all cases of malignant tumors. Compared to previous editions, the merits and demerits of the current revisions were analyzed. Many revisions have been made for criteria for the classification of lymph nodes. In particular, all the cases in whom the number of lymph nodes is more than 7 were classified as N3 without being differentiated. Therefore, the coverage of the N3 was broad. Owing to this, there was no consistency in predicting the prognosis of the N3 group. By determining the positive cases to a distant metastasis as TNM stage IV, the discrepancy in the TNM stage IV compared to the sixth edition was resolved. In regard to the classification system for an esophagogastric (EG) junction carcinoma, it was declared that cases of an invasion to the EG junction should follow the classification system for esophageal cancer. A review of clinical cases reported from Asian patients suggests that it would be more appropriate to follow the previous editions of the classification system for gastric cancer. In addition, in the classification of the TNM stages in the overall cases, the discrepancy in the prognosis between the different stages and the consistency in the prognosis between the same TNM stages were achieved to a lesser extent as compared to that previously. Accordingly, further revisions are needed to develop a purposive classification method where the prognosis can be predicted specifically to each variable and the mode of the overall classification can be simplified.

• Communications for Statistical Applications and Methods
• /
• 제26권6호
• /
• pp.635-644
• /
• 2019

Class probability is a fundamental target in classification that contains complete classification information. In this article, we propose a class probability estimation method when the predictor is functional. Motivated by Wang et al. (Biometrika, 95, 149-167, 2007), our estimator is obtained by training a sequence of functional weighted support vector machines (FWSVM) with different weights, which can be justified by the Fisher consistency of the hinge loss. The proposed method can be extended to multiclass classification via pairwise coupling proposed by Wu et al. (Journal of Machine Learning Research, 5, 975-1005, 2004). The use of FWSVM makes our method model-free as well as computationally efficient due to the piecewise linearity of the FWSVM solutions as functions of the weight. Numerical investigation to both synthetic and real data show the advantageous performance of the proposed method.

• 감성과학
• /
• 제26권3호
• /
• pp.15-28
• /
• 2023

본 연구는 정서 자극 유형 중 일상생활과 가장 유사한 정서 경험을 유발하는 자연주의적 자극인 영상 자극을 활용하여 정서표상의 유사성과 개인 간 반응 일관성을 살펴보기 위해 수행되었다. 이를 위해 다차원척도법을 실시하여 영상 자극이 핵심정서 차원에 위치하는지 확인하고, 참가자 간 분류분석을 사용하여 영상들이 정서유형 별로 구분이 잘 이루어지는지, 영상 자극에 대한 참가자들의 정서표상이 일관적인지 검증하였다. 또한 참가자간 상관분석을 통해 각 영상 자극에 대한 정서표상이 참가자들간 유사한지 추가적으로 확인하였다. 다차원척도법 결과, 정서유발 영상들이 정서가 차원에서 유의하게 구분되어 Russell(1980)의 핵심정서차원을 부분적으로 지지하였다. 분류분석 결과, 각 영상이 정서유형에 따라 잘 분류되고 예측되었다. 마지막으로, 참가자간 상관분석을 통해 정서 반응 일관성이 각 영상의 정서유형에 따라 다르게 나타남을 확인하였다. 본 연구는 영상 자극에 대한 정서표상과 정서 반응 일관성이 정서 유형에 따라 차이가 있음을 시사한다.

• 정보관리학회지
• /
• 제18권4호
• /
• pp.201-215
• /
• 2001

인터넷 쇼핑몰의 상품에 대한 분류체계를 도서관 분류이론에 적용하여, 효과적인 분류체계 구축을 위한 기준점을 파악하고자 하였다. 연구 방법은 기존의 웹 쇼핑몰 세 곳을 선정하여, 분류체계를 Ranganathan의 분류이론을 기준으로 하여 비교 분석하였다. 결과 크게 6가지 기준들이 파악되었는데, 상품의 특성, 범주의 포괄성, 다양한 접근점, 범부의 배열순서와 용어의 일관성, 용어의 최신성과 명백성, 용어의 반복적 사용의 금지들이었다. 추후 연구과제로 상품 탐색 형태와 인터페이스와의 관련성이 제시되었다.

• Korean journal of medical education
• /
• 제30권4호
• /
• pp.347-357
• /
• 2018

Purpose: The purpose of this study was to apply alternative standard setting methods for the Korean Medical Licensing Examination (KMLE), a criterion-referenced written examination, and to compare them to the conventional cut score used on the KMLE. Methods: The process and results of criterion-referenced standard settings (i.e., the modified-Angoff and bookmark methods) were evaluated. The ratio of passing and failing examinees determined using these alternative standard setting methods was compared to the results of the conventional criteria. Additionally, the external, internal and procedural evaluation of these methods were reviewed. Results: The modified-Angoff method yielded the highest cut score, followed sequentially by the conventional method and the bookmark method. The classification agreement between the modified-Angoff and bookmark methods was 0.720 measured by Cohen's ${\kappa}$ coefficient. The intra-panelist classification consistency of modified-Angoff method was higher than bookmark method. However, the inter-panelist classification consistency was vice versa. The standard setting panelists' survey results showed that the procedures of both methods were satisfactory, but panelists had more confidence in the results of the modified-Angoff method. Conclusion: The modified-Angoff method showed results that were more similar to those of the conventional method. Both new methods showed very high concordance with the conventional method, as well as with each other. The modified-Angoff method was considered feasible for adoption on the KMLE. The standard setting panelists responded positively to the modified-Angoff method in terms of its practical applicability, despite certain advantages of the bookmark method.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2000년도 추계학술대회 논문집 학회본부 D
• /
• pp.672-674
• /
• 2000

There are various ways in classification methodologies of data mining such as neural networks but the result should be explicit and understandable and the classification rules be short and clear. Rough set theory is a effective technique in extracting knowledge from incomplete and inconsistent information and makes an offer classification and approximation by various attributes with effect. This paper discusses granularity of knowledge for reasoning of uncertain concepts by using generalized rough set approximations based on hierarchical granulation structure and uses hierarchical classification methodology that is more effective technique for classification by applying core to upper level. The consistency rules with minimal attributes is discovered and applied to classifying real data.

• Steel and Composite Structures
• /
• 제7권3호
• /
• pp.185-200
• /
• 2007

Tubular construction is widely used in a range of civil and structural engineering applications. To date, the principal product range has comprised square, rectangular and circular hollow sections. However, hot-rolled structural steel elliptical hollow sections have been recently introduced and offer further choice to engineers and architects. Currently though, a lack of fundamental structural performance data and verified structural design guidance is inhibiting uptake. Of fundamental importance to structural metallic design is the concept of cross-section classification. This paper proposes slenderness parameters and a system of cross-section classification limits for elliptical hollow sections, developed on the basis of laboratory tests and numerical simulations. Four classes of cross-sections, namely Class 1 to 4 have been defined with limiting slenderness values. For the special case of elliptical hollow sections with an aspect ratio of unity, consistency with the slenderness limits for circular hollow sections in Eurocode 3 has been achieved. The proposed system of cross-section classification underpins the development of further design guidance for elliptical hollow sections.

• 대한원격탐사학회지
• /
• 제39권6_3호
• /
• pp.1771-1777
• /
• 2023

최근 딥러닝의 발전으로 의미론적 분할을 통한 토지피복 분류 방법들이 제안되고 있다. 그러나 기존의 딥러닝 기반 모델들은 영상 정보만을 이용하기 때문에 시공간적 일관성을 담보할 수 없는 한계점이 있다. 이에 본 연구에서는 좌표 정보를 활용한 토지피복 분류 모델을 제안한다. 먼저 암시적 신경 표현 기법인 다중해상도 해시 인코더를 위경도 좌표계로 확장한 좌표 해시 인코더를 통해 좌표의 특징을 추출하였다. 다음으로 추출된 좌표 특징을 다양한 단계의 U-net 디코더와 결합하는 아키텍처를 제안하였다. 실험 결과, 제안 방법이 약 32% 향상된 분류 정확도를 보였고, 시공간적 일관성이 향상됨을 확인하였다.