• 한국산림과학회지
• /
• 제99권5호
• /
• pp.736-743
• /
• 2010

This study examined the ecological attributes of classified forest community by cluster analysis in the mixed forest of Mt. Seokbyung Experimental Forest of Gangwon-Do. The vegetation data were collected in randomly established 51 sample plots (2.04 ha) and analysis adopted the cluster analysis, importance value index, and Shannon's diversity index. Main results were as follows; 1) the study area was classified into 4 clusters (A, B, C and D). 2) The cluster A was dominated by Pinus densiflora with an importance value of 71.6%. The most dominant species in the cluster B and cluster C were Larix leptolepis (57.1%) and Quercus mongolica (40.2%), respectively. Finally, The cluster D was dominated by P. densiflora (30.6%) and Q. mongolica (31.0%) with the mixed forest. 3) In the P. densiflora community (cluster A), distribution of DBH class showed a reverse J-shaped curve. In the L. leptolepis community (cluster B), individuals of dominant species had the bell-shaped distribution. Oak species indicated uniform distribution of DBH class (under 25 cm) in the mixed P. densiflora - Q. mongolica community (cluster D). 4) The species diversity index of the communities in descending order were: Pinus densiflora - Q. mongolica community > Larix leptolepis community > Pinus densiflora community > Quercus mongolica community.

• 환경영향평가
• /
• 제17권6호
• /
• pp.335-348
• /
• 2008

The purpose of this study is to classify the cities on sustainability assessment score studied in advance using cluster analysis, to present efficient management and policy direction based on analysis of sustainability index in 45 cities of all over Gyeongsangnam and Gyeongsangbuk-do. According to the results of cluster analysis, 45 cities were classed into 4 clusters by "livable-welfare city", "environmental -ecological city", "scientific-technological city", and "industrial-economic city". The livable-welfare cities must keep superior environmental sustainability, promote small and medium sized business on regional characteristic. The environmental-ecological cities have to change agriculture into future environmental industry such as ecotourism, bio-industry and landscape agriculture. The scientific-technological cities are going to need support of government scale such as income enlargement of citizen and stable job security. Finally, the industrial-economic cities must increase environmental management plants and improve quality of life through securing green spaces, maintaining public peace and applying UIS because of low quality of environment and life.

• Molecules and Cells
• /
• 제25권1호
• /
• pp.142-147
• /
• 2008

We recently used degenerate PCR and locus-specific PCR methods to identify the endogenous retroviruses (ERV) in the bovine genome. Using the ovine ERV classification system, the bovine ERVs (BERVs) could be classified into four families. Here, we searched the most recently released bovine genome database with the partial nucleotide sequence of the pro/pol region of the BERV ${\beta}3$ family. This allowed us to obtain and analyze the complete genome of BERV ${\beta}3$. The BERV ${\beta}3$ genome is 7666 nucleotides long and has the typical retroviral organization, namely, 5'-long terminal repeat (LTR)-gag-pro-pol-env-LTR-3'. The deduced open reading frames for gag, pro, pol and env of BERV ${\beta}3$ en- code 507, 271, 879 and 603 amino acids, respectively. BERV ${\beta}3$ showed little amino acid similarity to other betaretroviruses. Phylogenetic analysis showed that it clusters with HERV-K. This is the first report describing the genetic structure and sequence of an entire BERV.

• 대한가정학회지
• /
• 제34권2호
• /
• pp.183-199
• /
• 1996

The objective of this study was to provide fundamental data on somatotype for elderly women by classifying the somatotype and analysing the 3 characteristics of their somatotype. The subject were 368 women of 60~84 years old, they were analyzed indirect photography. To find out differences among the age groups, the 368 subjects were grouped into two age groups (Group 1 ; age 60 to 69, Group 2 ; aged 70 to 84). Data were analyzed using Factor analysis, Cluster analysis, duncan test and Analysis of variance. Through the factor analysis, 27 items from photometric measurements respectively. Cluster analysis was applied for classification of somatotype. The results of this study were as follows. 1. The characteristics of Elderly women's somatotype were bending of the upper-torso, fatness of the waist and abdomen, drooping of the bust and shoulder and hip. In addition, height, girth, depth and width items were decreased in their sizes respectively. 2. The characteristics of clusters were as follows. Type 1 was straight somatotype in which the plumb line through tragion, the bust depth and under bust depth region laterally. Type 2 was swayback somatotype in which the upper portion of protruding point on the back was bent forward but the lower portion of protruding point had a characteristics of turning over somatotype. Type 3 was bending somatotype. Namely, this type was shorter than average and below the average fatness. Generally, the lateral view silhouette of elderly women was the straight type and changed into bending type as the age increases.

• 한국군사과학기술학회지
• /
• 제19권3호
• /
• pp.346-355
• /
• 2016

In this paper, we propose a method to analyze the clusters of RF threats emitting electrical signals based on collected signal variables in integrated electronic warfare environments. We first analyze the signal variables collected by an electronic warfare receiver, and construct a model based on variables showing the properties of threats. To visualize the distribution of RF threats and reversely identify them, we use k-means clustering algorithm and self-organizing map (SOM) algorithm, which are belonging to unsupervised learning techniques. Through the resulting model compiled by k-means clustering and SOM algorithms, the RF threats can be classified into one of the distribution of RF threats. In an experiment, we measure the accuracy of classification results using the algorithms, and verify the resulting model that could be used to visually recognize the distribution of RF threats.

• Journal of Information Processing Systems
• /
• 제16권3호
• /
• pp.718-732
• /
• 2020

Conducting sentiment analysis and opinion mining are challenging tasks in natural language processing. Many of the sentiment analysis and opinion mining applications focus on product reviews, social media reviews, forums and microblogs whose reviews are topic-similar and opinion-rich. In this paper, we try to analyze the sentiments of sentences from online webcast reviews that scroll across the screen, which we call live barrages. Contrary to social media comments or product reviews, the topics in live barrages are more fragmented, and there are plenty of invalid comments that we must remove in the preprocessing phase. To extract evaluative sentiment sentences, we proposed a novel approach that clusters the barrages from the same commenter to solve the problem of scattering the information for each barrage. The method developed in this paper contains two subtasks: in the data preprocessing phase, we cluster the sentences from the same commenter and remove unavailable sentences; and we use a semi-supervised machine learning approach, the naïve Bayes algorithm, to analyze the sentiment of the barrage. According to our experimental results, this method shows that it performs well in analyzing the sentiment of online webcast barrages.

• 한국의류산업학회지
• /
• 제15권3호
• /
• pp.414-422
• /
• 2013

This study identifies the foot shapes of elderly men by classifying foot types according to the shapes of sole of foot and analyzing individual characteristics. The subjects were 269 elderly men over 60 years of age. Their right feet were measured indirectly with a 2D scanner. The anthropometric measuring items consisted of 38 items that were estimated on the right foot of each subject. The 2D scan data were analyzed by various statistical methods such as factor analysis, ANOVA and cluster analysis using the statistical program SPSS 19.0. A total of 8 factors were extracted through a factor analysis and these factors represent 77.83% of total variance. The 8 factors were: ball and lateral foot protrusion, ball gradient, medial foot protrusion, anterior and posterior foot length ratio, lateral ball length, heel size, toes breadth, and foot length, that explained 77.83% of the total variance. A total of 4 clusters (as their sole type) were categorized using 8 factor scores by cluster analysis. Type 1 was classified as H-type(toes width, foot width, heel width uniform and medial malleolus and lateral malleolus almost no protrusion). Type 2 was classified as V-type(foot width and toes width, wide and heel width narrow). Type 3 was classified as A-type(foot width and heel width, wide but toes width narrow, protruded inside). Type 4 was classified as D-type(protruded outside).

• Journal of Information Processing Systems
• /
• 제16권6호
• /
• pp.1238-1249
• /
• 2020

Throughout the world, aging populations and doctor shortages have helped drive the increasing demand for smart healthcare systems. Recently, these systems have benefited from the evolution of the Internet of Things (IoT), big data, and machine learning. However, these advances result in the generation of large amounts of data, making healthcare data analysis a major issue. These data have a number of complex properties such as high-dimensionality, irregularity, and sparsity, which makes efficient processing difficult to implement. These challenges are met by big data analytics. In this paper, we propose an innovative analytic framework for big healthcare data that are collected either from IoT wearable devices or from archived patient medical images. The proposed method would efficiently address the data heterogeneity problem using middleware between heterogeneous data sources and MapReduce Hadoop clusters. Furthermore, the proposed framework enables the use of both fog computing and cloud platforms to handle the problems faced through online and offline data processing, data storage, and data classification. Additionally, it guarantees robust and secure knowledge of patient medical data.

• 대한의용생체공학회:의공학회지
• /
• 제42권5호
• /
• pp.232-240
• /
• 2021

Heterogeneity in cancer is the major obstacle for precision medicine and has become a critical issue in the field of a cancer diagnosis. Many attempts were made to disentangle the complexity by molecular classification. However, multi-dimensional information from dynamic responses of cancer poses fundamental limitations on biomolecular marker-based conventional approaches. Cell morphology, which reflects the physiological state of the cell, can be used to track the temporal behavior of cancer cells conveniently. Here, we first present a hybrid learning-based platform that extracts cell morphology in a time-dependent manner using a deep convolutional neural network to incorporate multivariate data. Feature selection from more than 200 morphological features is conducted, which filters out less significant variables to enhance interpretation. Our platform then performs unsupervised clustering to unveil dynamic behavior patterns hidden from a high-dimensional dataset. As a result, we visualize morphology state-space by two-dimensional embedding as well as representative morphology clusters and trajectories. This cell morphology profiling strategy by hybrid learning enables simplification of the heterogeneous population of cancer.

• 한국대기환경학회지
• /
• 제17권4호
• /
• pp.331-345
• /
• 2001

The increase of carbon monoxide in the ambient air due to the emissions from point sources without control was estimated using three -dimensional numerical models. The target area was Ulsan where one of the largest industrial complexes was located. As a typical example using numerical models for air quality impact analysis of criteria pollutants that will determine whether the air quality standards would be exceeded or not, the following approaches were suggested. They include: (1) investigation of pre-existing atmospheric conditions, (2) identification of major factors causing high concentrations, (3) selection of episode days. (4) preparation of three-dimensional meteorological data, (5) confirmation of agreement between measured and predicted concentrations in the emission conditions of episode days, and (6) estimation of the impact due to changes of the emission conditions. In the present work, daily meteorological conditions for the specific period were classified into four clusters of distinctive features, and the episode days were selected individually from each cluster. Emphasis was placed on the selection of episodes representing meteorological conditions conducive to high concentrations especially for point sources that were sensitive to the wind direction variations.