• Journal of Soil and Groundwater Environment
• /
• v.12 no.5
• /
• pp.86-97
• /
• 2007

The principal component analysis was performed to identify the general characteristics of groundwater level changes from 202 deep and 112 shallow wells monitoring data, respectively, which came from the National Groundwater Monitoring Stations operated by KWATER with time spans of 156 continuous weeks from 2003 to 2005. Eight principal components, which accounted for 80% of the variability of the original time series, were extracted for water levels of shallow and deep monitoring wells. As a result of cluster analysis using the loading value of three principal components for shallow wells, shallow monitoring wells were divided into 3 groups which were characterized with a response time to rainfall (Group 1: 4.6 days, Group 2: 24.1 days, Group 3: 1.4 days), average long-term trend of water level (Group 1: $2.05{\times}10^{-4}$ m/day, Group 2: $-7.85{\times}10^{-4}$ m/day, Group 3: $-3.51{\times}10^{-5}$ m/day) and water level difference (Group 1 < Group 2 < Group 3). Additionally, they showed significant differences according to a distance to the nearest stream from well (Group 3 < Group 2 < Group 1), topographic slope of well site (Group 3: plain region, Group 1: mountainous region) and groundwater recharge rate (Group 3 < Group 2 < Group 1) with a p-value of 0.05.

• Korean Journal of Environment and Ecology
• /
• v.12 no.3
• /
• pp.212-223
• /
• 1998

The purpose of this study was to investigate the plant community structure and identify the change for 13 years(1983~1996) of forest around Youcheon industrial complex, Korea. 27 plots(300$m^2$/1plot) were established in forest around Youcheon industrial complex. By DCA ordination technique, the 27 plots were classified into five communities, which were Pinus thunbergii community, Pinus thunbergii-Quercus spp. community, P. thunbergii-Eurya japonica community, P. densiflora-P. thunbergii community and P. densiflora community. It seemed that P. thunbergii-Quercus app. community and P. densiflora-P. thunbergii community succeeded to Quercus spp. community and it seemed that P. thunbergii community, P. thunbergii-E. japonica community and P. densiflora community would not be replace by another woody species. Soil pH was pH 4.38~4.61, there were bad soil for organic matters content and exchangeable cations(C $a^{++}$, $Mg^{++}$, $K^{+}$) content. Shannon's diversity, H' max and number of species were improved for 13 years. So did soil characteristics.s.s.s.

• Proceedings of the Plant Resources Society of Korea Conference
• /
• 2019.10a
• /
• pp.84-84
• /
• 2019

본 연구는 전국 임의의 산양삼 재배지를 선정하여 재배지 내의 토양 특성 및 토양세균군집을 분석하고, 토양 특성, 세균군집 및 산양삼 생육특성 간의 상관관계를 구명하기 위하여 수행되었다. 토양 이화학성 분석은 농촌진흥청의 종합분석실 매뉴얼에 따라 분석하였고, 토양세균군집 분석은 pyrosequencing analysis (Illumina platform)를 이용하였다. 토양세균군집과 생육특성 간의 상관관계는 Spearman's rank correlation을 이용하여 분석하였다. 전국 8개 산양삼 재배지로부터 분리한 토양세균군집은 2개의 cluster로 군집화를 이루는 것을 확인하였다. 모든 토양 샘플에서 Proteobacteria와 Alphaproteobacteria가 각각 평균 상대적 빈도수가 35.4%, 24.4%로 우점종으로 나타났다. 나타났다. 두 개의 cluster 간 토양세균군집의 상대적 빈도수를 비교 분석한 결과, 먼저 Proteobacteria (p = 0.03), Actinobacteria (p = 0.02), Ahlpaproteobacteria (p = 0.029), Betaproteobacteria (p = 0.021)는 cluster 1에서 cluster 2에 비해 상대적 빈도수가 유의적으로 높았고, Fimicutes (p = 0.004), Cyanobacteria (p = 0.004), Acidobacteriia (p = 0.041), Ktedonobacteria (p = 0.019), Gammaproteobacteria (p = 0.034), Bacilli (p = 0.009)은 cluster 2에서 유의적으로 상대적 빈도수가 높은 것으로 나타났다. 토양세균군집 cluster 간 산양삼의 생육특성을 비교 분석한 결과, cluster 2 재배지에서 수집한 산양삼 시료의 지하부 생중량은 cluster 1 재배지에서 수집한 산양삼 시료에 비해 cluster 2에서 유의적 (p = 0.04)으로 높았다. 산양삼 생육특성과 토양세균군집 간의 상관관계를 분석한 결과, 산양삼의 생육은 토양 pH가 낮고 Acidobacteria의 상대적 빈도수가 높은 토양에서 증가하였으며, Acidobacteriia와 Koribacteraceae의 상대적 빈도수는 산양삼의 생육과 유의적인 정의 상관관계를 보이는 것으로 나타났다. 본 연구 결과는 토양미생물군집과 산양삼 생육 간의 상관관계를 구명하는 중요한 자료가 될 것으로 생각되고, 나아가 산양삼 재배적지를 선정하는데 있어 보다 명확한 정보를 제공할 수 있을 것으로 사료된다.

• The Korean Journal of Applied Statistics
• /
• v.33 no.5
• /
• pp.591-601
• /
• 2020

The self-organizing map (SOM) is a unsupervised learning method projecting high-dimensional data into low-dimensional nodes. It can visualize data in 2 or 3 dimensional space using the nodes and it is available to explore characteristics of data through the nodes. To understand the structure of data, cluster analysis is often used for nodes obtained from SOM. In cluster analysis, the optimal number of clusters is one of important issues. To help to determine it, various cluster validity indexes have been developed and they can be applied to clustering outcomes for nodes from SOM. However, while SOM has an advantage in that it reflects the topological properties of original data in the low-dimensional space, these indexes do not consider it. Thus, we propose a new cluster validity index for SOM based on connectivity between nodes which considers topological properties of data. The performance of the proposed index is evaluated through simulations and it is compared with various existing cluster validity indexes.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.04b
• /
• pp.247-249
• /
• 2002

협력적 여과(collaborative filtering) 방법을 사용하는 추천 시스템에서 예측 정확도를 높이는 방법들 중 하나는 군집화(clustering)방법이 있다. 군집화 방법은 선호도가 유사한 사용자들을 미리 같은 군집으로 만들고, 군집 내에 속한 사용자들을 이웃으로 선정하여 예측을 수행하기 때문에 군집화의 결과가 예측의 정확도에 직접적인 영향을 주게 된다. 본 연구에서는 군집화 결과의 향상을 위해 데이터를 전 처리하는 두 가지 방법과 군집화의 특성을 이용한 새로운 예측식을 제안하고, 기존 연구 방법과의 비교 실험을 통해 실험결과를 분석한다.

• Proceedings of the Korean Society of Disaster Information Conference
• /
• 2015.11a
• /
• pp.193-194
• /
• 2015

본 연구는 요인분석을 통해 노인 보행자 사고 유형을 대표할 수 있는 성분 값을 계산하고 군집분석을 실시하여 노인 보행자 사고 취약 지역을 선정하는 모델을 확인하였다. 기존 노인보행자 사고에 관한 연구는 보행 환경 분석 및 노인보행자 사고 특성을 확인한 뒤, 제도적, 물리적 환경 개선 등에 대한 정책 제언의 형식으로 진행되었다. 이러한 연구는 실질적으로 노인 보행자 특성을 분석하여 사고를 감소시키는 연구가 아닌, 일반적으로 알려진 사실이나 해외 사례를 들어 노인 보행자 사고의 현 실태에 대해 어느 정도 환기만 시켜주는 역할로 그쳤다. 이러한 점에서 군집분석을 통한 노인 보행자 사고 취약지역 선정은 노인 보행자 사고 특성의 비교를 명확하게 할 수 있도록 새로운 기준을 제시하였다. 이에 기존의 방법론에서 벗어나 실질적으로 노인 보행자 사고 방지 대책이 시급한 지역을 선정하였고, 노인 보행자 사고에 관해 활발히 연구 시킬 것이라 예상한다. 군집분석을 사용하는 핵심은 사망자 수와 사망률이 상대적으로 동시에 높은 군집을 선정할 수 있고 지역 특성 비교를 통해 향후 노인 보행자 사고에 관한 추가 연구가 가능할 것으로 기대한다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.435-435
• /
• 2023

최근 기후변화의 영향으로 상대적으로 좁은 국가면적이라 할 수 있는 국내에서도 가뭄과 홍수가 지역적으로 동시에 발생하는 등 자연재해에 의한 피해가 빈번해지고 있다. 이 중 가뭄은 타 풍수해에 비하여 일반적으로 장기간의 무강우에 의해 발현되고, 피해가 발생할 경우 생활용수, 농업용수, 공업용수 등 다양한 지역의 용수사용특성에 따라 피해가 집중 및 가중되는 특징을 지닌다. 본 연구에서는 지역별 가뭄피해도에 영향을 주는 영향인자의 민감도 분석을 실시하고 그 결과를 분석하였다. "가뭄 피해도"는 용수별 직간접 피해의 정도를 점수 또는 등급 등의 정량적으로 나타낸 것으로 정의된다. 우선, 국가가뭄정보통계집(2018-2020)에서 제시된 지역별 가뭄피해정보자료와 용수사용특성을 기반으로 군집분석을 실시한 문기훈(2022)의 연구모형을 기반으로 최근 발행된 2021년의 피해정보를 추가적으로 반영하여 군집분석을 실시하였다. 그리고 지역별 가뭄피해도 군집등급에 영향을 주는 과거 피해인자와 용수사용특성을 함수화하고, 지역별 불확실성을 반영한 민감도 분석을 실시하였다. 분석 결과 지역별 과거의 가뭄피해경험여부 및 용수사용특성의 변동에 따라 지역적 가뭄피해도 군집등급이 달라짐을 확인할 수 있었다.

• Korean Journal of Environmental Biology
• /
• v.37 no.3
• /
• pp.380-388
• /
• 2019

The studies regarding soil bacterial community and correlation analysis of wild-simulated ginseng cultivation area are insufficient. The purpose of this study was to investigate the correlation between soil bacterial community and growth characteristics of wild-simulated ginseng for selection of suitable cultivation area. The bacterial community was investigated by high throughput sequencing technique (Illumina platform). The correlation coefficient between soil bacterial community and growth characteristics were analyzed using Spearman's rank correlation. The soil bacterial community from soil samples of 8 different wild-simulated ginseng cultivated area exhibited two distinct clusters, cluster 1 and cluster 2. The relative abundance of Proteobacteria (35.4%) and Alphaproteobacteria(24.4%) was observed to be highest in all soil samples. The lower soil pH and higher abundance of Acidobacteria resulted in increased growth of wild-simulated ginseng. Additionally, abundance of Acidobacteriia (class) and Koribacteraceae (family) demonstrated significant positive correlation with fresh weight of wild-simulated ginseng. The results of this study clearly state the correlation between growth characteristic and soil bacterial community of wild-simulated ginseng cultivation area, thereby offering effective insight into selection of suitable cultivation area of wild-simulated ginseng.

• The Korean Journal of Applied Statistics
• /
• v.24 no.6
• /
• pp.1077-1094
• /
• 2011

We accomplish clustering analyses for yeast cell cycle microarray expression data. To reflect the characteristics of a time-course data, we screen the genes using the test statistics with Fourier coefficients applying a FDR procedure. We compare the results done by model-based clustering, K-means, PAM, SOM, hierarchical Ward method and Fuzzy method with the yeast data. As the validity measure for clustering results, connectivity, Dunn index and silhouette values are computed and compared. A biological interpretation with GO analysis is also included.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.16 no.4
• /
• pp.195-209
• /
• 2021

According to the 2019-2020 social media usage survey conducted by the Seoul e-commerce center, 5 out of 10 consumers have experienced shopping through social media. The cost of traditional advertising media has been reduced and advertising spending on social media has risen by 74%, indicating that social media is becoming a more important marketing element. While the number of users of social media has increased and corporate marketing activities have increased accordingly, research has been conducted in various aspects of marketing such as user motivation for social media, satisfaction, and purchase intention. There was no subdivided study on the differences in the social media usage frequency of consumers in actual purchasing behavior. This study attempted to identify differences in consumer characteristics by cluster in the agrifood purchase situation by grouping them by type according to the frequency of use of social media for consumers who purchase agri-food online. Product involvement, product need, and online purchase channel Consumer characteristics such as demographic distribution, perceived risk, and eating and lifestyle in each cluster were checked for the three agrifood purchase situations including choice, and types for each cluster were presented. To this end, questionnaire data on the frequency of social media use and online agrifood purchase behavior were collected from 245 consumers, and the validity of the measurement variables was secured through factor analysis and reliability analysis. As a result of cluster analysis according to the frequency of social media use, it was divided into three clusters. The first cluster was a group that mainly used open social media, and the second cluster was a group that used both open and closed social media and online shopping malls; The third cluster was a group with low online media usage overall, and the characteristics of each cluster appeared. Through regression analysis, the effect on product involvement, product need, and purchase channel selection when purchasing agri-food online through each of the three clusters was confirmed through regression analysis. As a result of the regression analysis, the characteristic of cluster 1 in the situation of purchasing agri-food online is a male in his 30s living in a rural area who has no reluctance to purchase agri-food on social media or online shopping malls. The characteristics of cluster 2 are mainly consumers who are interested in purchasing health food, and the consumer characteristics are represented. In the case of cluster 3, when purchasing products online, they purchase after considering quality and price a lot, and the consumer characteristics are represented as people who are more confident in purchasing offline than online. Through this study, it is judged that by identifying the differences in consumer characteristics that appear in the agri-food purchase situation according to the frequency of social media use, it can be helpful in strategic judgments in marketing practice on social media customer targeting and customer segmentation.