• The Korean Journal of Applied Statistics
• /
• v.33 no.2
• /
• pp.115-122
• /
• 2020

K-means clustering uses a spherical or elliptical metric to group data points; however, it does not work well for non-convex data such as the concentric circles. Spectral clustering, based on graph theory, is a generalized and robust technique to deal with non-standard type of data such as non-convex data. Results obtained by spectral clustering often outperform traditional clustering such as K-means. In this paper, we review spectral clustering and show important issues in spectral clustering such as determining the number of clusters K, estimation of scale parameter in the adjacency of two points, and the dimension reduction technique in clustering high-dimensional data.

• Land and Housing Review
• /
• v.10 no.1
• /
• pp.9-18
• /
• 2019

This study examined the dynamic change in the co-movement between the house price rates with the network methods of Mantegna (1999). First, Capital area and non-capital area form independent clusters which have the heterogeneous co-movement pattern. In other words, Capital and non-capital areas have low connectivity in the housing market. Also, if the co-movement between capital areas have been strengthened, the co-movement between non-capital areas have been weakened. The results of the dynamic analysis show that the degree of the co-movement in the housing market is continuously increased. The members of the co-movement group in the capital area are strongly steadied by all periods. However, the members in the non-capital area have been changed according to the period. Accordingly, it is necessary to establish policies based on various information for the housing market of the non-capital area rather than policies targeting the capital area. In addition, Apartments in Korea are more likely to be used as investment or speculative assets than other types of houses. It has been confirmed that this is Gangbuk, which is locatied in the northern part of Seoul, appears to be a region where the Spillover Effects of price fluctuation can be triggered in the housing and apartment market. However, the housing market in Gangnam, which is locatied in the southern part of Seoul, was divided into low systematic risk.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.6
• /
• pp.81-88
• /
• 2004

In this paper, we propose a new objective quality assessment method considering the human visual perception characteristics. A subjective quality assessment is obtained by the response of the receptive field in the primary visual cortex and a human's eye can't focus on all of the visual range in a moment. Take advantage of two facts above, we apply Gabor wavelet transform which is well fit the receptive field in the cortex, to divided constant sized subblocks. Then a local distortion of the subblocks and a global distortion for the entire image are calculated in order. The proposed method has been evaluated using video test sequences provided by the Video Quality Experts Group (VQEG). The experimental results show that good correlation with human perception is obtained using the proposed metric, which is what we called GPSNR.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.3
• /
• pp.976-990
• /
• 2020

The k-means algorithm is widely used in academia and industry due to easy and simple implementation, enabling fast learning for complex datasets. However, k-means struggles to classify datasets without prior knowledge of specific domains. We proposed the repulsive k-means (RK-means) algorithm in a previous study to improve the k-means algorithm, using the repulsive force concept, which allows deleting unnecessary cluster centroids. Accordingly, the RK-means enables to classifying of a dataset without domain knowledge. However, three main problems remain. The RK-means algorithm includes a cluster repulsive force offset, for clusters confined in other clusters, which can cause cluster locking; we were unable to prove RK-means provided optimal convergence in the previous study; and RK-means shown better performance only normalize term and weight. Therefore, this paper proposes the advanced RK-means (ARK-means) algorithm to resolve the RK-means problems. We establish an initialization strategy for deploying cluster centroids and define a metric for the ARK-means algorithm. Finally, we redefine the mass and normalize terms to close to the general dataset. We show ARK-means feasibility experimentally using blob and iris datasets. Experiment results verify the proposed ARK-means algorithm provides better performance than k-means, k'-means, and RK-means.

• Korean Journal of Soil Science and Fertilizer
• /
• v.26 no.3
• /
• pp.215-221
• /
• 1993

중국(中國)은 1991년에 새로운 토양계통분류법(土壤系統分類法)을 1차(次) 시안(試案)으로 확정 발표하였다. 본 체계는 미국의 Soil Taxonomy 체계와 비슷한 형태계량적규준(形態計量的規準)(Morpho-metric definition)으로 되어 있고 전산(電算)프로그램화(化)도 가능한 체계로 되어 있다. 분류체계는 목(目)(Order, 토망(土網)), 아목(亞目)(Suborder, 아망(亞網)), 군(群)(Group, 토류(土類)), 아군(亞群)(Subgroup, 아류(亞類)), 속(屬)(Family, 토속(土屬)), 통(統)(Series, 토종(土種))으로 되어 있고, 현재 13개 목망(目網), 33개 아목(亞目), 77개 군(群), 301개 아군(亞群)으로 분류되어 있다. 특징적표층(特徵的表層)은 유기표층(有機表層) 등(等) 8종, 진단표하층(診斷表下層)(차표층(次表層))은 표백층(漂白層) 등(等) 10종, 기타 특징토층(特徵土層)은 석회적표층(石灰的表層) 등(等) 12종, 그리고 암성(岩性) 등(等) 23종의 토양 생성적(生成的) 특징(特徵) 또는 현상(現像)을 분류기준으로 쓰고 있다. 분류상의 특징은 고위도(高緯度)나 고산지대(高山地帶)의 영구동결층(永久凍結層)을 강조하여 아목수준(亞目水準)에서 분류하였고 논토양이나 관개퇴적토(灌漑堆積土), 영양분(營養分)이 심층(深層)까지 집적(集積)된 채전(菜田) 등(等) 인위적(人爲的) 영향(影響)을 크게 받은 토양을 최고차(最高次) 분류단위(分類單位)인 목수준(目水準)에서 독립(獨立)시키고 있을 뿐 아니라 인위적인 교란(攪亂)을 받았으나 아직 단면(斷面)의 발달이 없는 토양을 초육토(初育土)의 아목수준(亞目水準)에서 구분하고 있다.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.43 no.5
• /
• pp.510-520
• /
• 2010

This study investigated the community structure and seasonal variation of the fish assemblages on Jangbong tidal flat, Incheon, Korea. Fish were collected monthly using a small otter trawl from November 1999 to January 2001. Thirty-six fish species were recorded, with a mean density of 185 individuals and biomass of 2,594.3 gWWt. The most abundant species by number were Johnius grypotus (23.7%), Acanthogobius hasta (17.8%), and Cynoglossus joyneri (10.7%), while the dominant species by catch weight were Acanthogobius hasta (21.2%), Sebastes schlegeli (16.2%), J. grypotus (14.0%), and C. joyneri (10.8%). Cluster analysis and non-metric multi-dimensional scaling (nMDS) were applied to assess the seasonal fluctuation in the fish assemblages. Based on the result of the cluster analysis and nMDS ordination, the faunal group could be divided into cold- and warm-water specialist groups. The cold-water specialists included A. hasta, Acanthogobius luridus, Triaenopogon barbatus, Tridentiger trigonocephalus, and Liza haematocheila. The warm-water specialists were J. grypotus, C. joyneri, S. schlegeli, and Hexagrammos otakii. In conclusion, the community structure showed a distinct seasonal trend, which seemed to be related to the seasonal fluctuations in water temperature.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.35 no.4
• /
• pp.16-23
• /
• 2012

At present, the significance of a new manufacturing system that can shift from 'mass production' and consider life cycles of a product is pointed out and extremely expected. In such a situation, it is recognized that the modular design, often called 'unit design,' is the important design methodology which realizes the new production system enabling 'cost reduction,' 'flexible production of a multi-functional artifact,' 'settlement of an environmental issue,' and so on. A module (unit) of a product is generally defined as 'the parts group made into the sub-system from a certain specific viewpoint.' So far, there have been many researches related to the modular design. However, they are often limited to a certain viewpoint (objective). This paper proposes a simple but effective method for multi-objective modular design. In the proposed method, a new design metric, called similarity index, is proposed to evaluate the modular design candidates from the multiple viewpoints.

• Journal of Digital Convergence
• /
• v.10 no.4
• /
• pp.199-206
• /
• 2012

Increment of mobile client's information request in wireless mobile networks requires a new method to manage and serve the streaming media object. This paper proposes a new object segment grouping method for enhancing the performance of streaming media services in wireless mobile networks. The proposed method performs the similarity metric for the partitioned object segments, and it process the disjunction, conjunction, and filtering for these metrics. This paper was to decided the partitioned group of object segments for these operation metrics, and it decided the performance of streaming media services. The simulation result showed that the proposed method has better performance in throughput, average startup latency, and cache hit ratio.

• Journal of the Korea Society of Computer and Information
• /
• v.24 no.7
• /
• pp.109-116
• /
• 2019

In this study, a GIS-based park safety assessment index was developed to enable objective park vulnerability assessment through data-based GIS analysis, a safety assessment was conducted by selecting a target site where various parks are operated and applying the park safety assessment index. In addition, a facility management system was developed for efficient management of the park safety assessment to update the park safety and provide a foothold for indirect PPGIS. In the case of the assessment index of the safety rating of the park, it was possible to conclude that the accurate quantitative performance was given to the calculation of the safety grade of the park based on the fact that the facilities are different depending on the environment and the size of the park. In addition, the marking the safety grade of parks, as well as the function to show the safety facilities of parks, a common living area for citizens, the management system is expected to have an impact on promoting the use of parks. In the future, in functions such as reporting of facility failures and verifying civil information are implemented by applying civic group participation programs and crowd-sourcing technologies, it is believed that all facilities as well as parks managed by the local government can be managed more efficiently.

• Ocean and Polar Research
• /
• v.33 no.3
• /
• pp.235-253
• /
• 2011

The annual variation of mesozooplankton community in the Jangmok harbor of Jinhae Bay was studied in relation to environmental variables. Sampling was carried out weekly from January to December 2009. During the study periods, mesozooplankton community consisted of 44 taxa and the annual mean abundance was 8308 inds. $m^{-3}$. The maximum abundance was observed to be 50043 inds. $m^{-3}$ in August and the minimum in April with 1013 inds. $m^{-3}$. Of these, Penilia avirostris, cirripedia larvae, Evadne tergestina, Acartia omorii, Oikopleura s, Paracalanus parvus s. l., Eurytemora pacifica, Podon s, Oithona s, and Acartia steueri were observed as dominant species in Jangmok bay and they also contributed to 79% of total mesozooplankton. According to non-metric multidimensional scaling (nMDS) and cluster analysis based on the mesozooplankton community data from each season, the community was divided into three groups. The first group included appearence species in winter and spring season, which is mainly dominated the copepod such as A. omorii and E. pacifica. The second and third group was composed with observed species in summer and autumn, respectively. Based on the SIMPER (similarity percentages), P. avirostris in summer and cirripedia larvae in autumn were significantly dominated. Our results indicate that although the mesozooplankton abundances in Jangmok harbor fluctuated abruptly, its annual variation was strongly influenced by water temperature.