• The Korean Journal of Applied Statistics
• /
• v.34 no.6
• /
• pp.905-922
• /
• 2021

Cluster analysis is one of unsupervised learning techniques used for discovering clusters when there is no prior knowledge of group membership. K-means, one of the commonly used cluster analysis techniques, may fail when the number of variables becomes large. In such high-dimensional cases, it is common to perform tandem analysis, K-means cluster analysis after reducing the number of variables using dimension reduction methods. However, there is no guarantee that the reduced dimension reveals the cluster structure properly. Principal component analysis may mask the structure of clusters, especially when there are large variances for variables that are not related to cluster structure. To overcome this, techniques that perform dimension reduction and cluster analysis simultaneously have been suggested. This study proposes probabilistic reduced K-means, the transition of reduced K-means (De Soete and Caroll, 1994) into a probabilistic framework. Simulation shows that the proposed method performs better than tandem clustering or clustering without any dimension reduction. When the number of the variables is larger than the number of samples in each cluster, probabilistic reduced K-means show better formation of clusters than non-probabilistic reduced K-means. In the application to a real data set, it revealed similar or better cluster structure compared to other methods.

• Journal of the Korean Society for information Management
• /
• v.29 no.2
• /
• pp.45-70
• /
• 2012

There have been many methods and algorithms proposed for multidimensional scaling to mapping the relationships between data objects into low dimensional space. But traditional techniques, such as PROXSCAL or ALSCAL, were found not effective for visualizing the proximities between objects and the structure of clusters of large data sets have more than 50 objects. The CLUSCAL(CLUster-oriented SCALing) technique introduced in this paper differs from them especially in that it uses cluster structure of input data set. The CLUSCAL procedure was tested and evaluated on two data sets, one is 50 authors co-citation data and the other is 85 words co-occurrence data. The results can be regarded as promising the usefulness of CLUSCAL method especially in identifying clusters on MDS maps.

• Korean Journal of Environment and Ecology
• /
• v.12 no.2
• /
• pp.138-146
• /
• 1998

Ten plots were established to study the vegetation structure of the Songinbong virgin forest in Ullungdo. It was classified into three communities Fagus crenata var. multinervis commnunity, Acer okamotoanum-Fagus crenata var. multinervis community, and Acer okamotoanum community. Shannon's diversity indicies of this area were not high(0.5150~0.8437), due to the geographical and vegetational characteristics. Considering the community structure(importance value, distribution of DBH class, and so on), the vegetation of the Songinbong virgin forest are thought to be stable continuously. With the analysis of plant growth type, it was presumed that Acer okamotoanum have increased after artificial disturbances in this area where Fagus crenata var. multinervis had been dominanat.

• Ecology and Resilient Infrastructure
• /
• v.2 no.1
• /
• pp.22-32
• /
• 2015

This study investigated the difference in fish community structures in a main channel and an isolated former channel, considering the environmental factors in the Mangyeong River, Korea. Principal component analysis (PCA) with environmental factors showed that former channels were composed of a fine substrate covered by in-stream vegetation, whereas the main channel was covered by a wide range of substrates with a higher dissolved oxygen and conductivity. The result of the hierarchical cluster analysis with species abundance delineated to the four main groups; three abandoned channel groups and one main channel group. Non-metric multidimensional scaling (NMDS) showed that fish community structures of each study site differed from environmental factors: former channel fish communities were positively related to in-stream vegetation cover, whereas main channel fish communities were positively associated with dissolved oxygen and conductivity. The results indicated that channelization, where there was a separation between the former channel and the main channel, had detrimental effects on fish community structures of both the main channel and the abandoned channel in the Mangyeong River. In conclusion, this study suggested that the connectivity between the main channel and abandoned channel were required to enhance both habitat structural diversity and species diversity of the Mangyeong River.

• Korean Journal of Environment and Ecology
• /
• v.8 no.2
• /
• pp.121-134
• /
• 1995

To investigate the plant community structure in Pinus densiflora forest of Chuwangsan National Park, twenty five plots were set up and surveryed. The classification by TWINSPAN and DCA ordination techniques were applied to the study area. The plant community were divided into five groups in twenty five plots by DCA and the dividing groups were Pinus densiflora-Quercus variabilis community(I), P. densiflora-Q. serrata community(II), P. densiflora-Fraxinus sieboldiana community(III), P. densiflora-F. sieboldiana-Q. spp. community(IV) and P. densiflora-mixed deciduous forest community(V). The successional trends of tree species by DCA ordination techniques and DBH class distribution analysis seems to be from P. densiflora through Q. mongolica, Q. serrata to Carpinus laxiflora in the canopy layer group. And in the subtree and shrub layer, it was expected that Lespedeza maximowiczii, L. cytobotrya, Rhododendron mucronulatumlongrightarrowStyrax obassia, Stephandra insisa, Zanthoxylum schinifolimlongrightarrowAcer pseudo-sieboldianum, Lindera obtusiloba. There was no difference between the stand scores of DCA and soil pH, and soil moisture.

• Korean Journal of Environment and Ecology
• /
• v.28 no.2
• /
• pp.197-204
• /
• 2014

The purpose of this study was to investigate the vegetation structure of Daphniphyllum macropodum community in the Yeong-sil and Seong-panack. To do so, 26plots ($(20m{\times}20m)$) were set up and surveyed. Plant communities were classified with 3 groups according to the analysis of classification by TWINSPAN and DCA Ordination; Pinus densiflora-Carpinus laxiflora (I), Carpinus laxiflora-Quercus serrata (II) and Carpinus tschonoskii (III). Daphniphyllum macropodum is appearing in understory and shrub layer of all groups. We found that Daphniphyllum macropodum will be dominated in understory and shrub layer. These results was judeged that these influence is depended on difference of altitude and accompanied species. Accompanied species were lived in similar growth environment to Daphniphyllum macropodum. That species were Styrax japonica, Quercus serrata of understory and Ilex crenata, Taxus cuspidata of shrub layer.

• Korean journal of applied entomology
• /
• v.49 no.1
• /
• pp.23-29
• /
• 2010

The environmental risks of cucumber mosaic virus resistant transgenic chili peppers with the CMVP0-CP gene on non-target organisms in the agroecosystem environments was evaluated during the periods of the chili pepper growing season (June 19, July 30, August 31) in 2007. Arthropods assemblages leaves and flowers of chili peppers were quantitatively collected by using an insect vacuum collector to compare the arthropod community structures between non-transgenic chili peppers (nTR, P 915) and mosaic virus resistant transgenic chili peppers (TR, CMV-cp, line 7). There was no statistical difference in the arthropod community structure between the two types of crops, nTR and TR, at the same season, although the species richness and Shannon's index were somewhat different among seasons; indicating no effects of genetically modified peppers on the arthropod community. However, further studies were required to conclude more concretely for the potential environmental risk of the transgenic chili pepper of CMV-cp.

• Korean Journal of Environment and Ecology
• /
• v.6 no.2
• /
• pp.147-153
• /
• 1993

The Dalbatjae-Birobong ridge forest in Mt. Sobaek area was studied to investigate forest community structure. Twenty 20m $\times$ 25m plots were set up in the studied ridge forest along altitude of 770m to 1,410m and vegetation analysis for tree stratum of the trees above 1cm DBH was carried out. The leading dominant tree species were Pinus densiflora in the lower elevation and Quercus mongolica in the higher elevation. According to importance value and cluster analysis, the studied ridge forest was classified into four forest communities of Pinus densiflora community, Pinus densiflora-Quercus mongolica community. Quercus mongolica community and Quercus mongolica-deciduous tree species community. According to importance value of major tree species in canopy and understory, the present community type seems to continue for a relatively long time. Species diversity of Pinus densiflora community, Pinus densiflora-Quercus mongolica community, Quercus mongolica community and Quercus mongolica-deciduous tree species community were 0.6791, 0.6850, 0.5296 and 0.7136, respectively.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.24 no.1
• /
• pp.9-16
• /
• 2011

An approach to understand the phase distribution in a multi-phase polycrystalline material is important since it can affect material properties and mechanical behaviors. A proper method is needed to describe the phase distribution. For this purpose, contiguity and probability functions(two-point correlation and lineal-path functions) are investigated for representing the phase distributions of microstructures. The mechanical behaviors are evaluated using the finite element method. The characteristics of probability functions and mechanical reponses of virtual samples are represented. It is confirmed that the topology of phase clustering affects the mechanical behavior of materials and that the strength is reduced as the clustering size increases.

• Korean Journal of Environment and Ecology
• /
• v.11 no.4
• /
• pp.433-449
• /
• 1998

To study the plant community structure of Chayang-chon area from Jangsudae to Hangyeryung in Soraksan National Park, 62 plots(each size 100m$^2$) were set up and surveyed. Six communities(Pinus densiflora - Populus caximowiczii community, P. densiflora community, P. densiflora - Deciduous broad-leaved tree community, Quercus mongolica community, Q. mongolica - Carpinus cordata community, C. cordata - Deciduous broad- leaved tree community) were classified by DCA ordination techniques. Q. mongolica community, Q. mongolica - C. cordata and C. cordata - Deciduous broad-leaved tree community were distrivuted over altitude 755m, and the others were distributed below there. In the community over altitude 755m, the succession that Q. mongolica community changed to C. cordata community have proceeded partly and otherwise C. cordata community and Deciduous broad-leaved tree community have been competing with each other. In the community with distribytion range from 500m to 560, high, P. densiflora community would change deciduous broad-leaved tree community forward.