• 한국어정보학회:학술대회논문집
• /
• 2016.10a
• /
• pp.157-161
• /
• 2016

본 논문에서는 클러스터 간의 중복을 허용한 계층적 클러스터링(hierarchical clustering) 기법에 적합한 클러스터 간 유사도 평가방법(linkage metric)을 제안하였다. 클러스터 간 유사도 평가방법은 계층적 클러스터링에서 클러스터를 통합하거나 분해하는데 쓰이며 사용된 방법에 따라 클러스터링의 결과가 다르게 형성된다. 기존의 클러스터 간 유사도 평가방법인 single linkage, complete linkage, average linkage 중 single linkage와 complete linkage는 클러스터 간 중복이 허용된 환경에서 정확도가 낮은 문제점이 있고, average linkage는 정확도가 두 방법에 비해 높지만 계산 시간 소요가 크다는 단점이 있다. 따라서 본 논문에서는 기존의 average linkage를 개선하여 중복된 데이터에 의한 필요 계산량을 크게 줄임으로써 시간적 성능이 우수한 클러스터 간 유사도 평가방법을 제안하였다. 또한, 제안된 방법을 기존 방법들과 비교실험하여 중복을 허용하는 계층적 클러스터링 환경에서 정확도는 비슷하거나 더 높고, average linkage에 비해 계산량이 감소됨을 확인하였다.

• Asian-Australasian Journal of Animal Sciences
• /
• v.8 no.6
• /
• pp.651-655
• /
• 1995

A simulation study on detection of linkage between genetic markers and QTL in backcross design was conducted. The effects of various sample sizes and the degree of QTL dominance on detention of linkage were examined by using a simple regression analysis. The results indicated that as sample size increased, the standard error of the estimated slope became smaller. When the dominance effect of QTL was complete, the estimated slope tended to be negative but was statistically not significant at all with type I error of greater than 50%. With complete linkage between genetic Marker and QTL, the estimated intercept value was smallest but the estimated slope was largest as expected. In most cases with various degree of dominance and sample sizes, when the actual recombination rate became larger, greater values were obtained for the slope except in the case of complete dominance of QTL.

• Annual Conference on Human and Language Technology
• /
• 2016.10a
• /
• pp.157-161
• /
• 2016

본 논문에서는 클러스터 간의 중복을 허용한 계층적 클러스터링(hierarchical clustering) 기법에 적합한 클러스터 간 유사도 평가방법(linkage metric)을 제안하였다. 클러스터 간 유사도 평가방법은 계층적 클러스터링에서 클러스터를 통합하거나 분해하는데 쓰이며 사용된 방법에 따라 클러스터링의 결과가 다르게 형성된다. 기존의 클러스터 간 유사도 평가방법인 single linkage, complete linkage, average linkage 중 single linkage와 complete linkage는 클러스터 간 중복이 허용된 환경에서 정확도가 낮은 문제점이 있고, average linkage는 정확도가 두 방법에 비해 높지만 계산 시간 소요가 크다는 단점이 있다. 따라서 본 논문에서는 기존의 average linkage를 개선하여 중복된 데이터에 의한 필요 계산량을 크게 줄임으로써 시간적 성능이 우수한 클러스터 간 유사도 평가방법을 제안하였다. 또한, 제안된 방법을 기존 방법들과 비교 실험하여 중복을 허용하는 계층적 클러스터링 환경에서 정확도는 비슷하거나 더 높고, average linkage에 비해 계산량이 감소됨을 확인하였다.

• Journal of the Korean Institute of Educational Facilities
• /
• v.15 no.3
• /
• pp.4-16
• /
• 2008

BTL(Build-Transfer-Lease) Project for Education Facilities is contracted as a package which consists of several education facilities and its maintenance period is 20 years. Thus, total cost variation largely depends on the accuracy of the maintenance cost forecasting in the early stage in the life cycle of the BTL Projects. This research develops a method using complete linkage algorithm and branch & bound algorithm to help in finding optimal bundling combination. The result of this research suggests more reasonable and effective forecasting method for the maintenance bundle in BTL projects.

• Journal of the Korean Society for information Management
• /
• v.21 no.1
• /
• pp.93-117
• /
• 2004

This study is to develop a hierarchic clustering model fur document classification and browsing in OPAC systems. Two automatic indexing techniques (with and without controlled terms), two term weighting methods (based on term frequency and binary weight), five similarity coefficients (Dice, Jaccard, Pearson, Cosine, and Squared Euclidean). and three hierarchic clustering algorithms (Between Average Linkage, Within Average Linkage, and Complete Linkage method) were tested on the document collection of 175 books and theses on library and information science. The best document clusters resulted from the Between Average Linkage or Complete Linkage method with Jaccard or Dice coefficient on the automatic indexing with controlled terms in binary vector. The clusters from Between Average Linkage with Jaccard has more likely decimal classification structure.

• Journal of the Korean Society for information Management
• /
• v.27 no.2
• /
• pp.241-252
• /
• 2010

Titles have been regarded as having effective clustering features, but they sometimes fail to represent the topic of a document and result in poorly generated document clusters. This study aims to improve the performance of document clustering with titles by suggesting titles in the citation bibliography as a clustering feature. Titles of original literature, titles in the citation bibliography, and an aggregation of both titles were adapted to measure the performance of clustering. Each feature was combined with three hierarchical clustering methods, within group average linkage, complete linkage, and Ward's method in the clustering experiment. The best practice case of this experiment was clustering document with features from both titles by within-groups average method.

• The Korean Journal of Applied Statistics
• /
• v.11 no.2
• /
• pp.233-246
• /
• 1998

We made a sample design for next 5 years, based on the population as of 1995, for livestock statistics. In the sample design, we used the stratified one stage sampling method where the sample size depends on the prefixed coefficient of variation. In stratifying the population, we considered the complete linkage method, and decided the number of strata to be the one which yields the minimum sample size. We listed here some difficulties we had for the better sample design in the future.

• Genomics & Informatics
• /
• v.7 no.1
• /
• pp.1-12
• /
• 2009

An alternative way of constructing ancestral graphs, which is different from the coalescent-based approach, is proposed using population linkage disequilibrium (LD) data. The main difference from the existing method is the construction of the ancestral graphs based on variants instead of individual sequences. Therefore, the key of the proposed method is to use the order of allele ages in the graphs. Distinct from the previous age-estimation methods, allele ages are estimated from full haplotype information by examining the number of generations from the initial complete LD to the current decayed state for each two variants depending on the direction of LD decay between variants. Using a simple algorithmic procedure, an ancestral graph can be derived from the expected allele ages and current LD decay status. This method is different in many ways from previous methods, and, with further improvement, it might be a good replacement for the current approaches.

• Honam Mathematical Journal
• /
• v.36 no.3
• /
• pp.605-622
• /
• 2014

We provide a minimal free resolution for a class of Gorenstein ideal of grade 4 which is the sum of an almost complete intersection J of grade 3 and a perfect ideal I of grade 3 with type 2 and ${\lambda}(I)$ > 0 geometrically linked by a regular sequence, where I is generated by odd elements.

• Honam Mathematical Journal
• /
• v.36 no.2
• /
• pp.387-398
• /
• 2014

In this paper, we give a structure theorem for a class of Gorenstein ideal of grade 4 which is the sum of an almost complete intersection of grade 3 and a Gorenstein ideal of grade 3 geometrically linked by a regular sequence. We also present the Hilbert function of a Gorenstein ideal of grade 4 induced by a Gorenstein matrix f.