• 한국지능시스템학회논문지
• /
• 제13권1호
• /
• pp.12-17
• /
• 2003

데이터의 군집화를 수행할 때 최적 군집수 결정은 군집 결과의 성능에 많은 영향을 미친다. 특히 K-means 방법에서는 초기 군집수 K에 따라 군집결과의 성능 차이가 많이 나타난다. 하지만 대다수의 군집분석에서 초기 군집수의 결정은 경험을 바탕으로 하여 주관적으로 결정된다. 이때 개체수와 속성수가 증가하면 이러한 결정은 더욱 어려워지며 이때 결정된 군집수가 최적이 된다는 보장도 없다. 본 논문에서는 군집의 수를 자동으로 결정하고 그 결과의 유효성을 보장하기 위해 유전자 알고리즘에 기반한 최적 군집수 결정 방안을 제안한다. 데이터의 속성에 근거한 초기 해 집단이 생성되고, 해 집단 내에서 최적화된 군집수를 찾기 위해 교차 연산이 이루어진다. 적합도 값은 전체 군집화의 비 유사성의 합의 역으로 결정되어 전체적인 군집화 성능이 향상되는 방향으로 수렴된다. 또한 지역 국소값을 해결하기 위해 돌연변이 연산이 사용된다. 그리고 유전자 알고리즘의 학습 시간의 비용을 줄이기 위해 붓스트랩 기법이 적용된다

• 한국발생생물학회지:발생과생식
• /
• 제19권4호
• /
• pp.259-264
• /
• 2015

Genomic DNA samples isolated from geographical purplish Washington clam (Saxidomus purpuratus) were obtained from three different regions in the Korean Peninsula: Geoje (Geoje population; GJP), Gunsan (Gunsan population; GSP) and a site of North Korea (North Korea population; NKP). The seven primers generated the total 369 loci that can be scored from the GSP clam population. 356 fragments were generated from the NKP clam population. The complexity of the banding patterns varies dramatically between the primers and three localities. In this study, 319 loci were identified in the purplish Washington clam from Geoje and 369 in the clam population from Gunsan: 221 specific loci (69.3%) in the GJP clam population and 300 (81.3%) in the GSP population. These results demonstrate that the primer detected a large quantity of specific fragments, suggesting that the genetic variation in the GSP is higher than in the GJP population. In particular, the BION-28 primer gave DNA profiles with more fragments than the other six primers in the NKP population. The oligonucleotides primer BION-75 produced 21 unique loci to each population, which were ascertaining each population, approximately 250 bp, 300 bp and 400 bp, in the GJP population. Outstandingly, the primer BION-50 detected 21 shared loci by the three populations, major and/or minor fragments of sizes 150 bp, which were matching in all samples. With regard to average bandsharing value (BS) results, individuals from GJP population (0.743) displayed higher bandsharing values than did individuals from GSP population (0.606). In the present study, the dendrogram gained by the seven oligonucleotides primers indicates three genetic clusters: cluster 1 (GEOJE 01 ~ GEOJE 07), cluster 2 (GUNSAN 08 ~ GUNSAN 14), cluster 3 (N.KOREA 15 ~ N.KOREA 21). Among the twenty one clams, the shortest genetic distance that revealed significant molecular differences was between individuals 08 and 09 from the NKP population (genetic distance = 0.073), while the longest genetic distance among the twenty-one individuals that demonstrated significant molecular differences was between individuals GEOJE no. 03 and GUNSAN no. 09 (genetic distance = 0.669). Comparatively, individuals of GJP population were properly closely related to that of NKP population, as revealed in the hierarchical dendrogram of genetic distances. In due course, PCR analysis has revealed the significant genetic distance among three purplish Washington clam populations. PCR fragments discovered in this study could be valuable as a DNA marker of the three geographical clam populations to distinguish.

• 제어로봇시스템학회논문지
• /
• 제7권12호
• /
• pp.985-992
• /
• 2001

This paper proposes a self-organizing fuzzy modeling(SOFUM)which an create a new hyperplane shaped cluster and adjust parameters of the fuzzy model in repetition. The suggested algorithm SOFUM is composed of four steps: coarse tuning. fine tuning cluster creation and optimization of learning rates. In the coarse tuning fuzzy C-regression model(FCRM) clustering and weighted recursive least squared (WRLS) algorithm are used and in the fine tuning gradient descent algorithm is used to adjust parameters of the fuzzy model precisely. In the cluster creation, a new hyperplane shaped cluster is created by applying multiple regression to input/output data with relatively large fuzzy entropy based on parameter tunings of fuzzy model. And learning rates are optimized by utilizing meiosis-genetic algorithm in the optimization of learning rates To check the effectiveness of the suggested algorithm two examples are examined and the performance of the identified fuzzy model is demonstrated via computer simulation.

• Journal of Information Technology Applications and Management
• /
• 제12권3호
• /
• pp.41-56
• /
• 2005

Product recommender system is one of the most popular techniques for customer relationship management. In addition, collaborative filtering (CF) has been known to be one of the most successful recommendation techniques in product recommender systems. However, CF has some limitations such as sparsity and scalability problems. This study proposes hybrid cluster analysis and case-based reasoning (CBR) to address these problems. CBR may relieve the sparsity problem because it recommends products using customer profile and transaction data, but it may still give rise to scalability problem. Thus, this study uses cluster analysis to reduce search space prior to CBR for scalability Problem. For cluster analysis, this study employs hybrid genetic and K-Means algorithms to avoid possibility of convergence in local minima of typical cluster analyses. This study also develops a Web-based prototype system to test the superiority of the proposed model.

• 농업과학연구
• /
• 제39권1호
• /
• pp.9-15
• /
• 2012

The study was conducted to evaluate the genetic similarity among commercial japonica rice varieties in Korea and China and to develop markers to differentiate between japonica cultivars developed in Korea and China. The genetic similarity and cluster of 38 accessions were analyzed using 47 SSR(simple sequence repeat) markers. The number of alleles by 47 SSR markers ranged from 2 to 9 with an average of 3.6. A total of 169 alleles were detected among these tested rice varieties. The PIC value varied from 0.05 to 0.79 with an average of 0.44. The Chinese japonica cultivars could be differentiated from the japonica cultivars in Korea by combining 2 SSR markers, RM223 and RM266. Cluster analysis showed that 38 tested varieties could be distinguished into japonica and indica based on the genetic distance.

• 정보처리학회논문지A
• /
• 제10A권1호
• /
• pp.15-24
• /
• 2003

본 논문에서는 분산 컴퓨팅 환경에서 클러스터 노드 할당 시스템에 대한 최적화 모델을 제시한다. 분산 파일 시스템 구조를 지닌 제시 모델에서는 시간에 따른 시스템의 역동적인 움직임을 면밀하게 고려하여 클러스터 노드 할당 세트가 타당한지를 조사하는 클러스터 모니터 노드의 기능이 주어진다. 노드 할당 시스템의 클러스터 모니터 노드는 병렬 모듈들을 클러스터 노드들에 분산시키면서 유전 알고리즘을 이용하여 좋은 할당 솔루션을 제공한다. 실험적 연구의 일환으로 코딩 기법, 교배, 돌연변이, 개체집단 크기 같은 다양한 유전 인자 파라미터와 노드 모듈개수에 따른 솔루션 품질 및 계산 시간에 관한 비교 실험 결과를 발표한다.

• 한국육종학회지
• /
• 제42권1호
• /
• pp.11-22
• /
• 2010

The population structure of a domesticated species is influenced by the natural history of the populations of its pre-domesticated ancestors, as well as by the breeding system and complexity of breeding practices implemented by humans. In the genetic and population structure analysis of 122 South Asia collections using 29 simple sequence repeat (SSR) markers, 362 alleles were detected, with an average of 12.5 per locus. The average expected heterozygosity and polymorphism information content (PIC) for each SSR locus were 0.74 and 0.72,respectively. The model-based structure analysis revealed the presence of three clusters with the 91.8% (shared > 75%) membership, with 8.2% showing admixture. The genetic distances of Clusters 1-3 were 0.55, 0.56, and 0.68, respectively. Polymorphic information content followed the same trend (Cluster 3 had the highest value and Cluster 1 had smallest value), with genetic distances for each cluster of 0.52, 0.52, and 0.65, respectively. This result could be used for supporting rice breeding programs in South Asia countries.

• Asian-Australasian Journal of Animal Sciences
• /
• 제32권4호
• /
• pp.467-476
• /
• 2019

Objective: This research was conducted to study the genetic diversity in several Indonesian cattle breeds using microsatellite markers to classify the Indonesian cattle breeds. Methods: A total of 229 DNA samples from of 10 cattle breeds were used in this study. The polymerase chain reaction process was conducted using 12 labeled primers. The size of allele was generated using the multiplex DNA fragment analysis. The POPGEN and CERVUS programs were used to obtain the observed number of alleles, effective number of alleles, observed heterozygosity value, expected heterozygosity value, allele frequency, genetic differentiation, the global heterozygote deficit among breeds, and the heterozygote deficit within the breed, gene flow, Hardy-Weinberg equilibrium, and polymorphism information content values. The MEGA program was used to generate a dendrogram that illustrates the relationship among cattle population. Bayesian clustering assignments were analyzed using STRUCTURE program. The GENETIX program was used to perform the correspondence factorial analysis (CFA). The GENALEX program was used to perform the principal coordinates analysis (PCoA) and analysis of molecular variance. The principal component analysis (PCA) was performed using adegenet package of R program. Results: A total of 862 alleles were detected in this study. The INRA23 allele 205 is a specific allele candidate for the Sumba Ongole cattle, while the allele 219 is a specific allele candidate for Ongole Grade. This study revealed a very close genetic relationship between the Ongole Grade and Sumba Ongole cattle and between the Madura and Pasundan cattle. The results from the CFA, PCoA, and PCA analysis in this study provide scientific evidence regarding the genetic relationship between Banteng and Bali cattle. According to the genetic relationship, the Pesisir cattle were classified as Bos indicus cattle. Conclusion: All identified alleles in this study were able to classify the cattle population into three clusters i.e. Bos taurus cluster (Simmental Purebred, Simmental Crossbred, and Holstein Friesian cattle); Bos indicus cluster (Sumba Ongole, Ongole Grade, Madura, Pasundan, and Pesisir cattle); and Bos javanicus cluster (Banteng and Bali cattle).

• Plant Resources
• /
• 제4권3호
• /
• pp.121-129
• /
• 2001

Intraspecific genetic diversity of sixteen accessions of Mogolian Alliums including fifteen species was investigated using randomly amplified polymorphic DNA (RAPD) analysis. Twenty three out of forty primers revealed scorable polymorphism. A total of 440 RAPD markers were generated on the 16 accessions of Mongolian Alliums. Among 440 RAPDs assayed, 439 were polymorphic with a mean polymorphic rate of 99.7%. Unweighted pair-group method using an arithmetic average (UPGMA) cluster analysis using RAPD data separated the 16 Allium accessions into two broad groups at similarity index 0.70. The clustering of the species was closely related with previous classification between A. altaicum and A. fistulosum. In addition, a high genetic similarity was showed between A. cepa and A. tagar.

• 한국발생생물학회지:발생과생식
• /
• 제10권1호
• /
• pp.19-32
• /
• 2006

참게(Eriocheir sinensis)와 꽃게(Portunus trituberculatus)의 2종으로부터 genomic DNA를 분리 추출하였다. 선택된 7개의 OPA-05, OPA-13, OPA-16, OPB-06, OPB-15, OPB-17 and OPD-10의 RAPD primer를 이용하여 identical, polymorphic 그리고 specific fragment를 얻어냈다. 본 연구에서 부안산 참게 집단에서는 505개의 fragment가 나타났고, 꽃게 집단에서는 513개의 fragment가 확인되었다. 참게 집단에서는 165개의 identical fragment가 나타났으며, 이는 primer당 평균적으로 23.6개의 fragment로 확인되었다. 또한 꽃게 집단에서는 66개로서 평균해서 primer당 9.4개의 identical fragment가 나타났다. 참게 집단과 꽃게 집단의 polymorphic fragment는 각각 50개와 14개로 나타났고, 참게 집단과 꽃게 집단의 경우 OPB-17에서 identical fragment가 300 bp의 크기에서 확인되었다. 각각을 비교해 보았을 때 유전적 차이는 참게 집단에서보다 꽃게 집단에서 더 높은 수치를 나타내었고, 2종 사이에서 $0.726{\pm}0.004$의 수치를 나타내었다. 7개의 primer를 사용하여 얻어진 dendrogram은 cluster 1(FRESHWATER 01), cluster 2(FRESHWATER 02, 03, 04, 05 및 06), cluster 3(FRESHWATER 07, 08, 09, 10 및 11) 및 cluster 4(SWIMMING 12, 13, 14, 15, 16, 17, 18, 19, 20, 21 and 22)와 같이 4개의 유전적 클러스터로 나뉘어졌다. 꽃게 집단에서 18번째 개체(SWIMMING no. 18)와 17번째 개체 (SWIMMING no. 17) 사이가 가장 가까운 유전적 관계(genetic distance 0.096)를 나타내었다. 궁극적으로 볼 때 참게 집단의 2번째(FRESHWATER no. 02)와 참게 집단의 3번째(FRESHWATER no. 03) 개체 사이가 가장 먼 유전적 거리 (genetic distance=0.770)를 나타내었다. 위에서 언급했던 것처럼 RAPD-PCR 방법은 참게 및 꽃게 2종의 종 판별을 하기 위한 진단적 표지 (diagnostic marker)로 이용할 수 있는 잠재력을 가지고 있는 것으로 확인되었다.