• Journal of the Korean Data and Information Science Society
• /
• v.10 no.1
• /
• pp.79-89
• /
• 1999

A counting rule that allows an element to be linked to more than one enumeration unit is called a multiplicity counting rule. Sample designs that use multiplicity counting rules are called network samples. Defining a network to be a set of observation units with a given linkage pattern, a network may be linked with more than one selection unit, and a single selection unit may be linked with more than one network. This paper considers allocation for multivariate stratified random sampling with multiplicity.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.918-919
• /
• 2023

딥러닝 기술이 발전함에 따라 모델의 복잡성 역시 증가하고 있다. 본 연구에서는 모델 경량화를 위해 Knowledge Distillation 기법을 Crowd Counting Model에 적용했다. M-SFANet을 Teacher 모델로, 파라미터수가 적은 MCNN 모델을 Student 모델로 채택해 Knowledge Distillation을 적용한 결과, 기존의 MCNN 모델보다 성능을 향상했다. 이는 정확도와 메모리 효율성 측면에서 많은 개선을 이루어 컴퓨팅 리소스가 부족한 기기에서도 본 모델을 실행할 수 있어 많은 활용이 가능할 것이다.

• Korean Journal of Optics and Photonics
• /
• v.9 no.3
• /
• pp.129-133
• /
• 1998

The phtocount distributions of a quasi-thermal light source were measured by using the photoelectric counting method. The source was fabricated from the ground glass plate of 9${\mu}{\textrm}{m}$ roughness illuminated by linearly polarized He-Ne laser beam. The coherence time of the quasi-thermal light changed from 31.4$mutextrm{s}$ to 2.48$mutextrm{s}$ according to the grain velocity of the ground glass plate. The photocount distribution showed the Bose-Einstein statistics for a long coherence time compared to the counting time interval, while the distribution approaches the Poisson distribution when the coherence time much shorter than the counting time.

• Journal of Korean Society for Quality Management
• /
• v.31 no.4
• /
• pp.219-225
• /
• 2003

Two­dimensional warranty data can be modelled using two different approaches： two­dimensional point process and one­dimensional point process with usage as a function of age. The first approach has three different models. First of all, bivariate model is appealing but is not appropriate for explaining warranty claims. Next, the rest of the two models (marked point process, and counting and matching on both directions independently) are more appropriate for explaining warranty claims. However, the second one (counting and matching on both directions independently) assumes that the two variables (variables representing the two­dimensions) are independent. Last of all, one­dimensional point process with usage as a function of age is also promising to explain the two­dimensional warranty claims. But the models or variations of them need more investigation to be applicable to real warranty claim data.

• Honam Mathematical Journal
• /
• v.32 no.1
• /
• pp.113-129
• /
• 2010

Inclusion and exclusion is used in many papers to count certain objects exactly or asymptotically. Also it is used to derive the Bonferroni inequalities in probabilistic area [6]. Inclusion and exclusion on finitely many types of properties is first used in R. Meyer [7] in probability form and first used in the paper of McKay, Palmer, Read and Robinson [8] as a form of counting version of inclusion and exclusion on two types of properties. In this paper, we provide a proof for inclusion and exclusion on finitely many types of properties in counting version. As an example, the asymptotic number of general cubic graphs via inclusion and exclusion formula is given for this generalization.

• Communications of the Korean Mathematical Society
• /
• v.39 no.1
• /
• pp.223-245
• /
• 2024

We study RNA foldings and investigate their topology using a combination of knot theory and embedded rigid vertex graphs. Knot theory has been helpful in modeling biomolecules, but classical knots emphasize a biomolecule's entanglement while ignoring their intrachain interactions. We remedy this by using stuck knots and links, which provide a way to emphasize both their entanglement and intrachain interactions. We first give a generating set of the oriented stuck Reidemeister moves for oriented stuck links. We then introduce an algebraic structure to axiomatize the oriented stuck Reidemeister moves. Using this algebraic structure, we define a coloring counting invariant of stuck links and provide explicit computations of the invariant. Lastly, we compute the counting invariant for arc diagrams of RNA foldings through the use of stuck link diagrams.

• Journal of the Korean Statistical Society
• /
• v.26 no.4
• /
• pp.429-443
• /
• 1997

We consider the problem of obtaining several types of simultaneous confidence bands for the survival curve under the additive risk model. The derivation uses the weak convergence of normalized cumulative hazard estimator to a mean zero Gaussian process whose distribution can be easily approxomated through simulation. The bands are illustrated by applying them from two well-known clinicla studies. Finally, simulation studies are carried outo to compare the performance of the proposed bands for the survival function under the additive risk model.

• Bulletin of the Korean Mathematical Society
• /
• v.54 no.6
• /
• pp.1883-1891
• /
• 2017

An optimally labelled graph of bandwidth 2 is an ordered pair (G, f) where G is a simple graph with bw(G) = 2 and $f:V(G){\rightarrow}[n]$ is a bijection such that bw(G, f) = 2. In this paper, the number of optimally labelled graphs of bandwidth two of order n is enumerated by counting linear forests.

• Journal of the Korean Mathematical Society
• /
• v.39 no.4
• /
• pp.495-507
• /
• 2002

In this paper, the characterization results related to Chernoff-type inequalities are applied for exponential-type (continuous and discrete) families. Upper variance bound is obtained here with a slightly different technique used in Alharbi and Shanbhag [1] and Mohtashami Borzadaran and Shanbhag [8]. Some results are shown with assuming measures such as non-atomic measure, atomic measure, Lebesgue measure and counting measure as special cases of Lebesgue-Stieltjes measure. Characterization results on power series distributions via Chernoff-type inequalities are corollaries to our results.

• Communications for Statistical Applications and Methods
• /
• v.5 no.1
• /
• pp.239-263
• /
• 1998

In this paper, we consider a multistate survival model which incorporates covariates and contains two illness states and two death states. The underlying stochastic process is assumed to follow nonhomogeneous Markov process. The estimates of survival, transition and competing risks probabilities are given via the methods of partial likelihood and nonparametric maximum likelihood. Our discussion is based on the statistical theory of counting process. An illustration is given to the data of patients in a heart transplant program. The goodness of fit procedures are also discussed to check the adequacy of the model.