• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권12호
• /
• pp.3815-3835
• /
• 2022

Density-based outlier detection is one of the hot issues in data mining. A point is determined as outlier on basis of the density of points near them. The existing density-based detection algorithms have high time complexity, in order to reduce the time complexity, a new outlier detection algorithm DODMD (Density-based Outlier Detection in Multidimensional Datasets) is proposed. Firstly, on the basis of ZH-tree, the concept of micro-cluster is introduced. Each leaf node is regarded as a micro-cluster, and the micro-cluster is calculated to achieve the purpose of batch filtering. In order to obtain n sets of approximate outliers quickly, a greedy method is used to calculate the boundary of LOF and mark the minimum value as LOF_min. Secondly, the outliers can filtered out by LOF_min, the real outliers are calculated, and then the result set is updated to make the boundary closer. Finally, the accuracy and efficiency of DODMD algorithm are verified on real dataset and synthetic dataset respectively.

• 한국멀티미디어학회논문지
• /
• 제12권11호
• /
• pp.1609-1622
• /
• 2009

무선 센서 네트워크에서 클러스터링 프로토콜은 전체 네트워크의 수명을 연장시키는 효율적인 방법이다. 그러나 클러스터 헤드 노드에 높은 부하를 주게 되어 헤드 노드의 급격한 에너지 소모를 유발하는 문제가 있다. 이에 LEACH와 같은 알고리즘은 클러스터의 구성과 클러스터 헤드 노드의 역할을 주기적으로 교체하여 네트워크의 수명을 연장시켰다. 그러나 이 방법은 클러스터를 구성하는 과정에서 상당한 양의 에너지를 소모한다. 이에 본 논문은 불필요한 에너지 소모를 줄이기 위해 새로운 클러스터 형성 알고리즘을 제안하였다. 이 알고리즘은 인접노드에서 수집되는 중첩 데이터를 배제하고 임계값을 전송한다. 서로 인접한 노드들은 그룹을 이루며 이 클러스터를 구성하는 노드들은 라운드 로빈 형태로 데이터를 수집하고 전송한다. 전체 네트워크의 관점에서 볼 때 이 그룹은 한개의 노드로 취급된다. 한 라운드의 셋업 단계에서 그룹들은 클러스터 헤드(그룹)에 의해 다시 클러스터를 형성(network cluster)하게 된다. 클러스터 헤드가 된 그룹의 모든 멤버노드는 라운드 로빈 방식으로 클러스터 헤드 역할을 수행한다. 따라서 그룹의 크기에 의해 라운드의 주기를 연장할 수 있다. 성능분석 결과 제안하는 방법은 제안된 클러스터링 방법에 비해 노드들의 에너지 소모가 줄어들었으며 전송효율이 증가하였다.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제29권4호
• /
• pp.248-257
• /
• 2002

군집화는 객체들의 특성을 분석하여 유사한 성질을 갖고 있는 객체들을 동일한 집단으로 분류하는 방법이다. 전자 상거래 자료처럼 차원 수가 많고 누락 값이 많은 자료의 경우 입력 자료의 차원축약, 잡음제거를 목적으로 SVD를 사용하여 군집화를 수행하는 것이 효과적이지만, SVD를 통해 변환된 자료는 원래의 속성 정보를 상실하기 때문에 군집 결과분석에서 원본 속성의 가치 해석이 어렵다. 따라서 본 연구는 군집화 수행 후 엔트로피 가중치 및 SVD를 이용하여 군집의 중요한 속성을 발견하기 위한 군집 특징 선택 기법 ENTROPY-SVD를 제안한다. ENTROPY-SVD는 자료의 속성들과 유사객체 군과의 묵시적인 은닉 구조를 활용하기 위하여 SVD를 이용하고 유사객체 군에 포함된 응집도가 높은 속성들을 발견하기 위하여 엔트로피 가중치를 사용한다. 또한 ENTROPY-SVD를 적용한 모델 기반의 협력적 여과기법의 추천 시스템 CFS-CF를 제안하고 그 효용성 및 효과를 평가한다.

• 한국가시화정보학회지
• /
• 제13권1호
• /
• pp.43-48
• /
• 2015

Collective cell migration is a fundamental phenomenon observed in various biological processes such as development, wound healing, and cancer metastasis. During the collective migration, cells undergo changes in their phenotypes from those of stable to the migratory state via the process called epithelial-mesenchymal transition (EMT). Recent findings in biology and biochemistry have shown that EMT is closely related to the cancer invasion or metastasis, but not much of the correlations in kinematics and physical forces between the neighboring cells are known yet. In this study, we aim to understand the cell migration and stress distribution within the expanding cell cluster. We constructed the in vitro cell cluster on the hydrogel, employed traction force microscopy (TFM) and monolayer stress microscopy (MSM) to visualize the physical forces within the expanding cell monolayer. During the expansion, cells at the cluster edge exhibited enhanced motility and developed focal adhesions that are the essential features of EMT while cells at the core of the cluster maintained the epithelial characteristics. In the aspect of mechanical stress, the cluster edge had the highest traction force of ~90 Pa directed toward the cluster core, which means that cells at the edge actively pull the substrate to make the cluster expansion. The cluster core of the tightly confined cells by neighboring cells had a lower traction force value (~60 Pa) but the highest intercellular normal stress of ~800 Pa because of the accumulation of traction from the edge of the monolayer.

• 한국환경복원기술학회지
• /
• 제26권6호
• /
• pp.51-65
• /
• 2023

The Jeongmaeks are Korea's unique forest space recognition system that diverged from the Baekdudaegan. The Jeongmaeks are easily exposed to pressure because it is adjacent to the living area. Among them, Nakdong-Jeongmaek has high biodiversity, but damage is accelerating. According to the Convention on Biological Diversity (CBD) in 2022, the target is to expand the area of terrestrial and marine protected areas to 30% of national territory by 2030. As of September 2023, the area of terrestrial protected areas in South Korea is only 16.97% of the country's territory. This is due in part to the high proportion of private forests in the region, which makes it difficult to establish protected areas. Therefore, there is a need to establish Other Effective Area-based Conservation Measure (OECMs), which pursue complex and effective conservation that considers multiple values, as an alternative to protected areas. This study aims to identify areas suitable for OECM and to provide opinions on the establishment of appropriate management plans for each value using SOM and InVEST Habitat Quality model. This study evaluated the habitat quality of 206 villages located within 1km of the Nakdong-Jeongmaek and compared the characteristics of villages classified by SOM. As a result, the habitat quality was 0.867 for Tourism village (ClusterIV), 0.838 for Conservation village (ClusterVI), 0.835 for Mixed village (ClusterI), 0.796 for Production (ClusterV), 0.731 for Rural village (ClusterIII) and 0.625 for Urban village (ClusterII). When the distribution was identified through statistical analysis, the Kruskal-Wallis test showed that the distributions were not identical, with a p-value of 1.53e-08. Dunn's test showed a difference between Tourism, Conservation and Rural, Urban village. However, Mixed village was overestimated due to the lack of villages and the small area included in the study area. Moreover, Conservation village was somewhat under-evaluated in the analysis due to the use of a single weight for protected areas. It is necessary to perform additional reinforcement of the value evaluation of Jeongmaeks by conducting Forest Resource Survey and the National Natural Environment Survey. Therefore, we believe that sufficient validity for the establishment of OECMs in the Nakdong-Jeongmaek can be provided by addressing these limitations and conducting additional research.

• 한국의류산업학회지
• /
• 제18권5호
• /
• pp.606-616
• /
• 2016

We intend an empirical assessment of examining the differences in the appearance management behaviors and demographic variables among groups classified by the clothing consumption values. The questionnaires are administered to 493 female and male adults above 20 years old in Seoul, Gyeonggi-do, Daegu and Kyungpook regions. For analysis of data from 478 respondents, descriptive statistics, cluster analysis, Cronbach's ${\alpha}$, ANOVA, Duncan test and ${\chi}^2$ test were applied. We show the following results. First, Factor analyses were employed for the clothing consumption values and appearance management behaviors. Six factors were for clothing consumption values: Individuality, appearance attractive, social, functional, conditional and fashion clothing consumption value. Four factors were for appearance management behaviors: weight training, skin care, hair care, make-up and clothing selection. According to clothing consumption values, four groups were classified: the passive, functional, social, and active group. We did cluster analysis to the appearance management behaviors of weight training, skin care, hair care, make-up and clothing selection. Second, the social and active groups were more interested in individuality, appearance attractive, social, functional, conditional and fashion clothing value. And they were also more involved in appearance management behaviors. Third, among the demographic variables, the single and female in 20s and 30s with higher level of education belonged to the active group. In this contribution, we find significant differences in the appearance management behavior and demographic variables classified by the clothing consumption values.

• 한국경제지리학회지
• /
• 제17권3호
• /
• pp.465-481
• /
• 2014

창의산업(creative industry)으로서 의류패션산업은 지역에 착근된 가치사슬을 바탕으로 한 기업 간 협력과 상호 네트워크가 중요한 산업으로, 특정지역을 중심으로 집적하고 있다. 이에 본 연구는 공간적 집적을 넘어 서울 의류패션산업의 중심 클러스터로 성장하고 있는 구로 가산동 의류패션산업의 가치사슬을 바탕으로 산업구조를 분석하고 기업간 산업네트워크를 형성하고 있는지 확인하였다. GIS분석과 현장답사를 통해 구로 가산동 의류패션산업은 가치사슬상의 기획, 디자인의 고부가가치 영역이 전개되고 있음을 확인하였다. 또한, 심층인터뷰를 통해 구로 가산동 의류패션산업은 가치사슬의 미들스트림을 중심으로 업스트림에 속하는 지역 외부의 대형 브랜드업체의 통제를 받는 외부지향적인 생산구조를 가지고 있음을 밝혔다. 이에 구로 가산동의 산업 집적지는 상호호혜적인 네트워크를 통한 창조적인 산업클러스터로 발전하는데 한계가 존재한다. 하지만, 부가가치가 높은 업스트림의 영역으로 가치무리 이동을 보이고 있어, 현재 일부 기업들이 부가가치가 높은 영역을 넓혀 가고 있는 모습이 존재함을 보였다. 본 연구는 구로 가산동 의류패션산업의 총체적인 시스템을 확인하고, 경제지리적 관점에서 가치사슬의 구조와 공간적 분화패턴을 국지적 차원에서 분석한데에 의미가 있다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권6호
• /
• pp.2644-2657
• /
• 2018

To utilize the energy of sensor nodes efficiently and extend the network lifetime maximally is one of the primary goals in wireless sensor networks (WSNs). Thus, designing an energy-efficient protocol to optimize the determination of cluster heads (CHs) in WSNs has become increasingly important. In this paper, we propose a novel energy-efficient protocol based on an improved Grey Wolf Optimizer (GWO), which we refer to as Fitness value based Improved GWO (FIGWO). It considers a fitness value to improve the finding of the optimal solution in GWO, which ensures a better distribution of CHs and a more balanced cluster structure. According to the distance to the CHs and the BS, sensor nodes' transmission distance are recalculated to reduce the energy consumption. Simulation results demonstrate that the proposed approach can prolong the stability period of the network in comparison to other algorithms, namely by 31.5% in comparison to SEP, and even by 57.8% when compared with LEACH protocol. The results also show that the proposed protocol performs well over the above comparative protocols in terms of energy consumption and network throughput.

• 농업과학연구
• /
• 제45권3호
• /
• pp.547-559
• /
• 2018

Consumers are becoming increasingly interested in what kind of value their food has. Many studies have focused on consumers' preferences and willingness to pay for specific food values. However, few studies have asked consumers to consider or rank the importance of different food values. This paper determined consumers' food values by implementing the best-worst scaling approach and segmented consumers based on the relative importance of general food values that consumers place on them. Among a list of eleven food values (taste, safety, origin, appearance, price, environmental impact, naturalness, convenience, nutrition, fairness, and habit) which was compiled from previous studies on food preferences, on average, safety, nutrition, taste, and price were the most important values to consumers, whereas fairness, habit, appearance, convenience, origin, and environmental impact were the least important values. However, significant variation exists among consumers in terms of the relative importance of food values. To investigate the heterogeneity among consumers, a Latent Class Analysis was performed to classify consumers into subgroups based on responses to questions. Two latent classes were found and characterized as 'safety-nutrition' and 'taste-price'. The 'safety-nutrition' cluster represents 61% of the sample and a group of people who find safety and nutrition centered values to be the most important. Another cluster represents about 39% of the sample, and relative to the first group, this group finds price and taste values to be more important.

• Journal of Electrical Engineering and Technology
• /
• 제13권6호
• /
• pp.2468-2478
• /
• 2018

We propose the Sparse Feature Convolutional Neural Network (SFCNN) to reduce the volume of convolutional neural networks (CNNs). Despite the superior classification performance of CNNs, their enormous network volume requires high computational cost and long processing time, making real-time applications such as online-training difficult. We propose an advanced network that reduces the volume of conventional CNNs by producing a region-based sparse feature map. To produce the sparse feature map, two complementary region-based value extraction methods, cluster max extraction and local value extraction, are proposed. Cluster max is selected as the main function based on experimental results. To evaluate SFCNN, we conduct an experiment with two conventional CNNs. The network trains 59 times faster and tests 81 times faster than the VGG network, with a 1.2% loss of accuracy in multi-class classification using the Caltech101 dataset. In vehicle classification using the GTI Vehicle Image Database, the network trains 88 times faster and tests 94 times faster than the conventional CNNs, with a 0.1% loss of accuracy.