• 한국데이터정보과학회:학술대회논문집
• /
• 2005.10a
• /
• pp.47-48
• /
• 2005

Multidimensional scaling is a multivariate technique for constructing a configuration of n points in Euclidean space using information about the distances between the objects. This can be done by the singular value decomposition of the data matrix. But it is known that the singular value decomposition is not resistant. In this study, we provide a resistant version of the multidimensional scaling.

• Journal of Korean Society for Quality Management
• /
• v.32 no.3
• /
• pp.68-80
• /
• 2004

This paper purports to uncover the underlying attributes used by customers to gauge service quality. Data was collected by administering questionnaires to 50 respondents and then analyzed by using Multidimensional Scaling methodology. The findings indicate that there are two primary dimensions to service quality. This analysis helped determine us two alternatives to naming the dimensions. Experience properties of service and Price value of the service, or Responsiveness of service provider employees and Reliability of service providers.

• Proceedings of the Korean Society for Quality Management Conference
• /
• 2004.04a
• /
• pp.639-645
• /
• 2004

This paper purports to uncover the underlying attributes used by customers to gauge service quality. Data was collected by administering questionnaires to 50 respondents and then analyzed by using Multidimensional Scaling methodology. The findings indicate that there are two primary dimensions to service quality. A considerable analysis helped determine two alternatives to naming the dimensions: Experience properties of service and Price value of the service, or Responsiveness of service provider employees and Reliability of service providers.

• Journal of the Korean Society for information Management
• /
• v.29 no.2
• /
• pp.45-70
• /
• 2012

There have been many methods and algorithms proposed for multidimensional scaling to mapping the relationships between data objects into low dimensional space. But traditional techniques, such as PROXSCAL or ALSCAL, were found not effective for visualizing the proximities between objects and the structure of clusters of large data sets have more than 50 objects. The CLUSCAL(CLUster-oriented SCALing) technique introduced in this paper differs from them especially in that it uses cluster structure of input data set. The CLUSCAL procedure was tested and evaluated on two data sets, one is 50 authors co-citation data and the other is 85 words co-occurrence data. The results can be regarded as promising the usefulness of CLUSCAL method especially in identifying clusters on MDS maps.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.34 no.2
• /
• pp.60-67
• /
• 2011

Since the non-metric Multidimensional scaling (nMDS), a data visualization technique, provides with insights about engineering, economic, and scientific applications, it is widely used for analyzing large non-metric multidimensional data sets. The nMDS requires a fitness function to measure fit of the proximity data by the distances among n objects. Most commonly used fitness functions are nonlinear and have a difficulty to find a good configuration. In this paper, we propose a new fitness function, an absolute value type, and show its advantages.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.4
• /
• pp.427-431
• /
• 2010

Multidimensional scaling (MDS) is a widely used method for dimensionality reduction, of which purpose is to represent high-dimensional data in a low-dimensional space while preserving distances among objects as much as possible. MDS has mainly been applied to data visualization and feature selection. Among various MDS methods, the classical MDS is not readily applicable to data which has large numbers of objects, on normal desktop computers due to its computational complexity. More precisely, it needs to solve eigenpair problems on dissimilarity matrices based on Euclidean distance. Thus, running time and required memory of the classical MDS highly increase as n (the number of objects) grows up, restricting its use in large-scale domains. In this paper, we propose an efficient approximation algorithm for the classical MDS based on divide-and-conquer and CUDA. Through a set of experiments, we show that our approach is highly efficient and effective for analysis and visualization of data consisting of several thousands of objects.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.14 no.2
• /
• pp.129-136
• /
• 1990

The purpose of the study was to develope brand positioning map for ladies' ready-to-wear, to find out evaluative criteria in perception and preference to brands, and to persent the relationship between consumer's characteristics and brand preference. Subjects were selected for the housewives of middle and high socioeconomic classes living in Seoul area. A questionnaire including items of life style, self image, similarity between brands, preference degree to brands, and demographic variables was developed for the empirical study. The questionnaire was administrated to 137 housewives during fall in 1989. Data were analyzed by cluster analysis and multidimensional scaling method. The study had two research problems. The first research problem was to construct a brand perceptual map for ladies' ready-to-wear brands, selected for the study The perceptual map was constructed on the basis of brand similarity scores by multidimensional scaling method. As a result, brands were grouped into 4 clusters, and evaluative criteria for perceptual map were found to be fashionability (classic- fashionable) and familiarity (familiar-unfamiliar). The second problem was to construct a brand preference map for ladies' ready-to-wear brands, selected for the study. The preference map was constructed on the basis of brand preference scores by multidimensional scaling method. As a result, the brands were grouped into 4 clusters and evaluative critiera for preference map were found to be fashionability (unfashionable-fashionable) and image to age (mature-young directed). Also was shown the relationship among self image, age, socioeconomic class, and brand preference. The multidimensional scaling method was found to be useful as well as valid instrument for brand positioning research and the result can be utilized for establishing strategies for ladies' ready-to-wear brands.

• The Korean Journal of Applied Statistics
• /
• v.27 no.4
• /
• pp.619-627
• /
• 2014

Distances or dissimilarities among units are assumed to be symmetric in most cases of multidimensional scaling(MDS); consequently, it is not an easy task to deal with asymmetric distances. Current asymmetric MDS still face difficulties in the interpretation of results. This study proposes a simpler asymmetric MDS that utilizes the order statistic of an asymmetric matrix. The proposed Web method demonstrates that some influences among objects are visualized by direction, size and shape of arrow to ease the interpretability of users.

• The Korean Journal of Applied Statistics
• /
• v.25 no.4
• /
• pp.613-620
• /
• 2012

In most cases of multidimensional scaling(MDS), the distances or dissimilarities among units are assumed to be symmetric. Thus, it is not an easy task to deal with asymmetric distances. Asymmetric MDS developed so far face difficulties in the interpretation of results. This study proposes a much simpler asymmetric MDS, that utilizes the notion of "altitude". The analogy arises in mountaineering: It is easier (more difficult) to move from the higher (lower) point to the lower (higher). The idea is formulated as a quantification problem, in which the disparity of distances is maximally related to the altitude difference. The proposed method is demonstrated in three examples, in which the altitudes are visualized by rainbow colors to ease the interpretability of users.

• The Journal of Korea Robotics Society
• /
• v.3 no.2
• /
• pp.117-122
• /
• 2008

This paper presents a multi-robot localization based on multidimensional scaling (MDS) in spite of the existence of incomplete and noisy data. While the traditional algorithms for MDS work on the full-rank distance matrix, there might be many missing data in the real world due to occlusions. Moreover, it has no considerations to dealing with the uncertainty due to noisy observations. We propose a robust MDS to handle both the incomplete and noisy data, which is applied to solve the multi-robot localization problem. To deal with the incomplete data, we use the Nystr$\ddot{o}$m approximation which approximates the full distance matrix. To deal with the uncertainty, we formulate a Bayesian framework for MDS which finds the posterior of coordinates of objects by means of statistical inference. We not only verify the performance of MDS-based multi-robot localization by computer simulations, but also implement a real world localization of multi-robot team. Using extensive empirical results, we show that the accuracy of the proposed method is almost similar to that of Monte Carlo Localization(MCL).