• 디지털융복합연구
• /
• 제17권4호
• /
• pp.355-364
• /
• 2019

본 연구는 관광SNS 러브마크경험을 근간으로 이용자의 시장세분화 및 특성을 분석하여 시사점을 제시하고자 하였다. 연구의 결과는 다음과 같다. 첫째 관광SNS 러브마크경험에 대한 요인 분석 결과 신비감, 감각성, 친밀감, 신뢰, 평판의 5가지 요인을 도출하였다. 둘째, 군집분석 결과 평판/감각성중심경험집단, 고 러브마크경험집단, 친밀감중심경험집단, 저 러브마크경험집단으로 이루어지는 4개의 군집을 추출하였으며 각 군집별 특징에 대한 상세적 분석을 진행하였다. 셋째, 교차분석과 대응일치분석을 실시한 결과 군집별로 관광SNS 이용 특성에 차이가 있는 것으로 나타났다. 넷째, 관광SNS 러브마크경험을 높게 인식하는 집단일수록 경험적 가치, 만족, 행동의도를 높게 인식하는 것으로 나타났다.

• 한국의류산업학회지
• /
• 제13권3호
• /
• pp.364-372
• /
• 2011

The primary purposes of this study were to classify consumer's type according to narcissistic orientation of consumer and then to reveal consumer's perception differences on consumption value of apparel products and cosmetics subject to the types. Data were collected from 194 female college students in Ulsan. It was analyzed by frequency analysis, factor analysis, cluster analysis and MANOVA, ANOVA using SPSS Win 12.0. The results were as follows: First, consumer's types classified according to narcissistic orientation were non-narcissist, high-narcissist, leadership and self-absorption. Second, in the consumption value of apparel products case, there were significant differences in effective value, symbolic value and sensitive value among the types. Third, in the consumption value of cosmetics case, there were significant differences in sensitive value and effective value among the types.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권4호
• /
• pp.1773-1795
• /
• 2020

Unmanned aerial vehicles (UAVs) network are a very vibrant research area nowadays. They have many military and civil applications. Limited bandwidth, the high mobility and secure communication of micro UAVs represent their three main problems. In this paper, we try to address these problems by means of secure clustering, and a security clustering algorithm based on integrated trust value for UAVs network is proposed. First, an improved the k-means++ algorithm is presented to determine the optimal number of clusters by the network bandwidth parameter, which ensures the optimal use of network bandwidth. Second, we considered variables representing the link expiration time to improve node clustering, and used the integrated trust value to rapidly detect malicious nodes and establish a head list. Node clustering reduce impact of high mobility and head list enhance the security of clustering algorithm. Finally, combined the remaining energy ratio, relative mobility, and the relative degrees of the nodes to select the best cluster head. The results of a simulation showed that the proposed clustering algorithm incurred a smaller computational load and higher network security.

• 한국생활과학회지
• /
• 제7권2호
• /
• pp.63-73
• /
• 1998

To get well-fitted ready-made clothings with beautiful silhouettes, it's better to classify body forms into several forms and to assign sizing within each form than to grade just based on body size regardless of body styles. This study illucidated the importance of drop value in the results of surveying the current values of sizing and grading. Therefore, it's meaningful to get the classification of body form with appropriate distribution of drop values of the body, and the distribution of drop value and the frequency of each form is very helpful to name the combined sizing or coverage of ready-made clothes. This study aimed at classifying body forms with various drop values using multivariate analysis for sizing and grading. Factor analysis and cluster analysis were done using measured values from 346 unmarried women. The results are as follows: 1. The factor which explains body forms was obtained by factor analysis, and the representative major 18 items which have important roles in classifying body forms were selected among the measured values with high factor loading and communality. 2. The body forms were classified into 8 groups based on the charateristics, frequencies and distributions of them obtained from cluster analysis. 3. Each classified body form showed conspicuous difference in drop value and the difference of body form mainly resulted from the difference between waist and hip rather than the difference between bust circumference and waist in Korean unmarried women.

• 동아시아식생활학회지
• /
• 제20권6호
• /
• pp.987-996
• /
• 2010

The purpose of this study was to measure customers' perception of the service quality of Korean restaurants and then compare differences in perceived service quality according to customer nationality. Self-administered questionnaires were completed by 2812 subjects, and data were analyzed by frequency, chi-square, t-test, one-way ANOVA, factor, reliability, cluster, and discriminant analysis. Results of the study were as follows. The factor analysis of perceived service quality produced four factors, employee service (5 variables), menu quality (4 variables), price & value (4 variables), and physical environment (4 variables). Cronbach's alpha values for reliability were over 0.8 for all factors. Further, a significant difference was observed in service quality, which was perceived according to customer nationality. A higher mean value of perceived service quality was held by foreigners when compared to Koreans. Especially, the mean value of perceived service quality was significantly low for all items for Japanese compared to foreigners. Cluster analysis divided subjects into two groups based on attitude toward service quality of Korean restaurants: an unfavorable group and favorable group. These two groups differed from each other in general characteristics as well. Limitations and future research directions are also discussed.

• 지능정보연구
• /
• 제21권2호
• /
• pp.93-112
• /
• 2015

소비자의 소비성향이 필요 품목을 중심으로 근거리에서 구매하는 근린형으로 변화함에 기존의 소매점은 식료품, 생활용품을 위주로 제공하는 슈퍼마켓, 하이퍼마켓 또는 편의점으로 진화하고 있다. 따라서 소매점이 한정된 공간에서 효율적으로 공간을 활용하고 매출을 증대하기 위해서는 소비자의 구매욕을 충족시킬 수 있는 상품배치와 적정한 재고수준을 유지하는 것이 매우 중요하다. 본 연구에서는 소매점의 판매 상품에 대하여 RFM 기반 SOM 군집화를 하여 효율적으로 매장을 관리할 수 있는 상품 배치전략 및 재고전략을 제안하였다. 실제 M마트의 판매데이터를 이용하여 RFM모델을 상품에 적용한 후, 기존 문헌 연구뿐만 아니라 해석 가능성, 응용 가능성 등을 고려하여 3X3 총 9개의 군집으로 분류하여 분석한 결과, 주요 군집으로 R값, F값, M값이 모두 높은 군집, R값, F값, M값 모두 낮은 군집, R값만 높은 군집, F값만 높은 군집이 도출되었다. 본 논문에서는 다른 군집과 비교시 R값, F값, M값이 차이를 보이는 주요 4개의 군집의 상품 배치 및 재고 전략을 제시하였다. R값, F값, M값이 모두 높은 군집의 상품은 소비자 동선을 늘림으로써 상품 노출을 확대시킬 수 있는 장소에 배치하여야 할 뿐만 아니라 높은 수준의 재고를 보유할 필요가 있다. 반면에 R값, F값, M값이 모두 낮은 군집의 상품은 가시성이 낮은 곳에 배치하고 최소한의 안전재고만 보유할 필요가 있다. 또한 R값이 높은 군집은 신상품으로 매장 입구에 배치하여 상품의 판매를 유도할 필요가 있다. 그리고 F값만 높은 군집의 경우, R값과 M값이 평균 값 보다 작은 상품들의 군집이므로 최근에는 판매가 저조하며 빈도 수에 비해 총 판매액이 낮다는 것을 유추할 수 있다. 따라서 현재보다 과거에 많이 판매된 저가의 상품군집으로 재고 수준을 점차 감소시킬 필요가 있다. 본 연구에서 제시한 방법은 POS 시스템의 보유한 소매점에서 상품배치 및 재고관리 방법으로 활용되어 매장의 수익성 증대에 기여할 수 있을 것으로 기대된다.

• Environmental Engineering Research
• /
• 제15권3호
• /
• pp.135-140
• /
• 2010

With the various urban characteristics of each city, the existing water demand prediction, which uses average liter per capita day, cannot be used to achieve an accurate prediction as it fails to consider several variables. Thus, this study considered social and industrial factors of 164 local cities, in addition to population and other directly influential factors, and used main substance and cluster analyses to develop a more efficient water demand prediction model that considers unique localities of each city. After clustering, a multiple regression model was developed that proved that the $R^2$ value of the inclusive multiple regression model was 0.59; whereas, those of Clusters A and B were 0.62 and 0.74, respectively. Thus, the multiple regression model was considered more reasonable and valid than the inclusive multiple regression model. In summary, the water demand prediction model using principal component and cluster analyses as the standards to classify localities has a better modification coefficient than that of the inclusive multiple regression model, which does not consider localities.

• 천문학회지
• /
• 제32권2호
• /
• pp.91-107
• /
• 1999

We present BVRI CCD photometry and near-infrared K photometry of the intermediate-aged open cluster NGC 7790. The reddening, E(B - V) = 0.54 $\pm$ 0.05 and the distance modulus, (m - M)o = 12.45 $\pm$ 0.10 for the cluster were determined by zero-age-main-sequence fitting and theoretical isochrone fitting using not only (V, B - V), (V, V - 1), (V, V - R) but also (V, V - K) color-magnitude diagrams. The reddening corresponded approximately to the average value derived from previous studies, while the distance modulus was found to be almost midway between the CCD photometric results of Romeo et al. (1989) and those of Mateo & Madore (1988). We have used four colors to distinguish members from field stars. The expected colors were calculated using the derived distance modulus, and were then were compared with the observed colors (B - V), (V - 1), (V - R), and (V - K). Thus, a color excess E(B - V) for each star was determined which could give the minimum difference between the calculated and observed colors. Single and binary members of the cluster were determined on the basis of the E(B - V) distribution of stars.

• 한국발생생물학회지:발생과생식
• /
• 제16권4호
• /
• pp.315-320
• /
• 2012

The gDNA isolated from Cyclina sinensis from Gochang (GOCHANG), Incheon (INCHEON) and a Chinese site (CHINESE), were amplified by PCR. Here, the seven oligonucleotide decamer primers (BION-66, BION-68, BION-72, BION-73, BION-74, BION-76, and BION-80) were used to generate the unique shared loci to each population and shared loci by the three cyclina clam populations. As regards multiple comparisons of average bandsharing value results, cyclina clam population from Chinese (0.763) exhibited higher bandsharing values than did clam from Incheon (0.681). In this study, the dendrogram obtained by the seven decamer primers indicates three genetic clusters: cluster 1 (GOCHANG 01~GOCHANG 07), cluster 2 (INCHEON 08~INCHEON 14), cluster 3 (CHINESE 15~CHINESE 21). The shortest genetic distance that displayed significant molecular differences was between individuals 15 and 17 from the Chinese cyclina clam (0.049), while the longest genetic distance among the twenty-one cyclina clams that displayed significant molecular differences was between individuals GOCHANG no. 03 and INCHEON no. 12 (0.575). Individuals of Incheon cyclina clam population was somewhat closely related to that of Chinese cyclina clam population. In conclusion, our PCR analysis revealed a significant genetic distance among the three cyclina clam populations.

• 정보관리학회지
• /
• 제21권4호
• /
• pp.173-185
• /
• 2004

정보통신의 기술이 발달하면서 정보의 양이 많아지고 사용자의 질의에 대한 검색 결과 리스트도 많이 추출되므로 빠르고 고품질의 문서 클러스터링 알고리즘이 중요한 역할을 하고 있다. 많은 논문들이 계층적 클러스터링 방법을 이용하여 좋은 성능을 보이지만 시간이 많이 소요된다. 반면 K-means 알고리즘은 시간 복잡도를 줄일 수 있는 방법이다. 본 논문에서는 계층적 클러스터링 시스템인 콘도르(Condor) 시스템에서 간단하고 고품질이며 효율적으로 정보 검색 할 수 있도록 구현하였다. 이 시스템은 K-Means Algorithm을 이용하였으며 클러스터 계층 깊이와 초기값을 조절하여 $88\%$의 정확율을 보였다.