• The Korean Journal of Applied Statistics
• /
• v.31 no.1
• /
• pp.67-75
• /
• 2018

Multidimensional scaling (MDS) is an exploratory analysis of multivariate data to represent the dissimilarity among objects in the geometric low-dimensional space. However, a general MDS map only shows the information of objects without any information about variables. In this study, we used MDS based on the algorithm of Torgerson (Theory and Methods of Scaling, Wiley, 1958) to visualize some clusters of objects in categorical data. For this, we convert given data into a multiple indicator matrix. Additionally, we added the information of levels for each categorical variable on the MDS map by applying the partition method of Shin et al. (Korean Journal of Applied Statistics, 28, 1171-1180, 2015). Therefore, we can find information on the similarity among objects as well as find associations among categorical variables using the proposed MDS map.

• The KIPS Transactions:PartD
• /
• v.16D no.5
• /
• pp.683-690
• /
• 2009

Social Networks which is composed of network with an individual in the center in a web support mutual-understanding of information by searching user profile and forming new link. Therefore, if we apply the Social Network which consists of web users who have similar immanent information to web search, we can improve efficiency of web search and satisfaction of web user about search results. In this paper, first, we make a Social Network using web users linked directly or indirectly. Next, we calculate Similarity among web users using their immanent information according to topics, and then reconstruct Social Network based on varying Similarity according to topics. Last, we compare Similarity with Search Pattern. As a result of this test, we can confirm a result that among users who have high relationship index, that is, who have strong link strength according to personal attributes have similar search pattern. If such fact is applied to search algorithm, it can be possible to improve search efficiency and reliability in personalized and social search.

• Annual Conference of KIPS
• /
• 2013.11a
• /
• pp.1193-1195
• /
• 2013

최근 정보 기술의 발달로 XML, 화학 복합물, 소셜 네트워크 등과 같은 구조적 정보를 갖는 빅 데이터들이 대량으로 축적되고 있다. 이러한 구조적 정보를 갖는 그래프 데이터에서 메디안을 찾기 위한 연구가 진행되고 있다. 기존에는 그래프 메디안을 효율적으로 계산하기 위해 하한값을 이용한 그래프 메디안 탐색 기법이 제안되었다. 그러나 탐색을 시작하는 버텍스를 선정하는 데에 따라 가지치기 효과가 다르게 발생하는 문제점이 있다. 본 논문에서는 버텍스의 그래프 구조 정보를 이용한 효율적인 메디안 탐색 기법을 제안한다. 제안하는 탐색 기법은 버텍스의 차수와 에지 가중치를 이용하여 그래프 메디안 예측 값을 정의하고, 그래프 메디안과 유사한 버텍스들부터 우선적으로 탐색한다. 실험을 통하여 제안하는 기법이 기존의 방법보다 최대 10%까지 수행시간을 단축함을 보인다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.6A
• /
• pp.584-591
• /
• 2002

In this paper, we proposed a b1ocking artifact reduction algorithm using similarity and linear combination between blocks. In the proposed method, all of the blocks are classified into low frequency block and high frequency block according to the their DCT coefficients. And we defined range block which shows blocking artifacts, as block centered to block boundary and defied domain block as similar block with range block within search range. In the search procedure, we used sub-block's property of range block and similarity for more accurate searching. Finally blocking artifact reduction algorithm is performed using linear combination between searched domain block and blocky range block. The performance of the proposed method is investigated by computer simulation in comparison with the traditional methods. In the experimental results, Ire confirmed the better performance in the subjective by 0.04∼0.4 dB and objective image quality.

• Annual Conference of KIPS
• /
• 2016.10a
• /
• pp.663-664
• /
• 2016

다양한 분야에서 특정 이미지와 유사한 내용의 이미지를 찾기 위해서 이미지 유사도 기법을 지원하는 웹 또는 프로그램이 많이 사용되고 있다. 본 논문에서는 이미지 유사도 기법을 병리학 이미지에 적용하여 특정 세포를 찾는데 사용함으로써 사용자에게 정확하고 신뢰성 있는 정보를 전달하려 한다. 실제로 병리학에서 특정 세포를 찾기에는 세포의 크기가 너무 작고 모양이 일정하지 않는다는 점 때문에 유사한 세포를 찾기가 힘들다. 이를 해결하기 위해 Leveling과 Tiling 방식을 적용하여 데이터를 추출하고 이미지 동기화를 통해 유사도 비교 시 정확도를 높이도록 한다. 다양한 병리학 이미지 포맷을 지원하는 OpenSlide 기반의 대용량 이미지에 제안하는 방식을 적용하여 특정 세포와 유사한 부분을 찾아내는 실험을 통하여 제안하는 방식이 효율적임을 보인다.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.43 no.6 s.312
• /
• pp.57-64
• /
• 2006

Motion estimation plays an important role for the compression of video signals. The proposed method utilizes an adaptive threshold and characteristics of a distribution of SAD (sum of absolute difference). Generally, the more complex the SAD distribution is, the larger SAD value tends to be. This proposed algorithm tries to reduce the search points in a simple distribution but increase them in a complex distribution to avoid local minima. A macro block is divided into 9 areas. One of them chosen using spatio-temporal correlation is called the primary area and the others are called the secondary area that will be searched to avoid local minima. The proposed algorithm decides if just one area (the primary area or the secondary area) will be enough to be searched or both areas should be searched, using adaptive threshold. Compared with famous motion estimation algorithms, the simulation result shows that the searching points per macro block and MSE decreases about 16.4% and 32.83 respectively on the average.

• Korean journal of communication and information
• /
• v.36
• /
• pp.374-398
• /
• 2006

This paper attempts to identify motivations of the satellite DMB use and to examine the relationship between user motivations and structure of genre consumption. Firstly, this paper identified motivational dimensions for satellite DMB usage. The results of the factor analysis revealed three dimensions: information-seeking, entertainment-oriented and social communication. It accounts for 69.9% of the variances found within this data. Secondly, the results of the regression analyses suggested that both information seeking and entertainment-oriented motivations were closely related to genre preferences of satellite DMB contents. In addition, these two motivations predicted genre repertoires and similarity in genre consumption. In conclusion, motivations of the satellite DMB use were the best predictors for explaining the structure of genre consumption.

• Journal of Global Scholars of Marketing Science
• /
• v.13
• /
• pp.49-70
• /
• 2004

Recently, consumer researchers have suggested that consumers evaluate brand by subjective memory-based-emotion which was formed by their experience. However, relatively little consumer research has been done to explore conditions under which memory-based-emotion is used in brand evaluation. Therefore, this study explored the conditions to facilitate usage of memory-based- emotion such as accessibility, representativeness, relevance of consumption goal, when consumers evaluate brand. In addition, we investigated the factors that influence responses of memory-based- emotion, such as similarity of encoding and retrieving condition and level of organization. The results of this study suggest that level of organization was positively related to accessibility and representativeness. But similarity of encoding and retrieving condition was not related to them. And accessibility, representativeness influence on use of memory-based-emotion in consumer brand evaluation. But relevance of consumption goal did not influence on use of memory-based-emotion in consumer brand evaluation.

• Journal of Information Management
• /
• v.26 no.2
• /
• pp.1-36
• /
• 1995

The purpose of this study is to analyze fundamental six functions of DIALOG, STN, ORBIT and CDP and then investigate the similarities of the four systems in terms of the functions, and also to compare the commands of these systems with ISO 8777 which is an international standard of command language for the interactive text searching. It was found that there are few differences between the four systems and ISO 8777 in functional aspects, but few similarities in command expression aspects. In addition, it was proven that these four systems are more effective than ISO 8777 in some commands.

• Applied Biological Chemistry
• /
• v.51 no.3
• /
• pp.219-222
• /
• 2008

In the search for third generation herbicidal cyclic imide derivatives, the average values of herbicidal activity ($pI_{50}$) in vivo (pre-emergence) of 40 new peroxidizing herbicides, 1-(4-chloro-2-fluoro-5-propargyloxyphenyl)thiourea derivatives (1-40) against rice plant (Orysa sativa) and barnyard grass (Echinochlor crus-galli) were studied. The molecular similarity between protoporphyrinogen IX (protogen) as the substrate of protox enzyme and Urea derivatives (1-40) was discussed quantitatively. The diallyl (20) and 3-nitro substituent (33) showed the selective herbicidal activity against barnyard grass. Allyl substituent (8) and their molecular similarity in dice (S=0.81) showed the highest levels of herbicidal activity ($pI_{50}$=4.71). Also, similarity indices (S) and superimposed volume (C) of protogen and aryl-substituents (21-40) showed good correlation.