The Relationship between Multidimensional Scaling and Clustering

1977 ◽  
pp. 17-44 ◽  
Author(s):  
Joseph Kruskal
Keyword(s):  
Multidimensional Scaling ◽  
The Relationship

scholarly journals The bimodal character of stock distribution

2001 ◽  
Vol 32 (3) ◽  
pp. 11-16
Author(s):  
A. G. Frank
Keyword(s):  
Life Cycle ◽  
Multidimensional Scaling ◽  
Monthly Basis ◽  
Corporate Life Cycle ◽  
Time Changes ◽  
Degree Of Differentiation ◽  
Relationship Of ◽  
The Relationship ◽  
Over Time ◽  
Growth Probability

The bimodal character of stocks is demonstrated when they are classified according to their style. Stocks are often assigned, on the basis of some valuation parameter, uniquely as either value or growth even though, over time, changes in a stock’s-growth probability should trace the evolution of the corporate life cycle. This study is concerned with investigating the relationship of that probability to market cycles. Two hundred and eighty eight stocks from the ASEAN are tracked over an eight-year period. The percentages of those, on a monthly basis, that are in the top quintile of EPS growth, as well as the top quintile of major value (current) styles are calculated. Using multidimensional scaling, the study concludes that the degree of differentiation between growth and value rises as the market declines, and that styles are purer at the bottom than at the top of the market cycle.

scholarly journals An Interlanguage Unification of Musical Timbre

2015 ◽  
Vol 32 (4) ◽  
pp. 394-412 ◽  
Author(s):  
Asterios Zacharakis ◽  
Konstantinos Pastiadis ◽  
Joshua D. Reiss
Keyword(s):  
Multidimensional Scaling ◽  
Energy Distribution ◽  
Acoustic Analysis ◽  
Perceptual Information ◽  
Scaling Analysis ◽  
Strong Correlations ◽  
Multidimensional Scaling Analysis ◽  
Acoustic Correlates ◽  
Linguistic Groups ◽  
The Relationship

The current study expands our previous work on interlanguage musical timbre semantics by examining the relationship between semantics and perception of timbre. Following Zacharakis, Pastiadis, and Reiss (2014), a pairwise dissimilarity listening test involving participants from two separate linguistic groups (Greek and English) was conducted. Subsequent multidimensional scaling analysis produced a 3D perceptual timbre space for each language. The comparison between perceptual spaces suggested that timbre perception is unaffected by native language. Additionally, comparisons between semantic and perceptual spaces revealed substantial similarities which suggest that verbal descriptions can convey a considerable amount of perceptual information. The previously determined semantic labels “auditory texture” and “luminance” featured the highest associations with perceptual dimensions for both languages. “Auditory mass” failed to show any strong correlations. Acoustic analysis identified energy distribution of harmonic partials, spectral detail, temporal/spectrotemporal characteristics and the fundamental frequency as the most salient acoustic correlates of perceptual dimensions.

Neural Feature Abstraction from Judgments of Similarity

1998 ◽  
Vol 10 (7) ◽  
pp. 1815-1830 ◽  
Author(s):  
Michael D. Lee
Keyword(s):  
Neural Network ◽  
Multidimensional Scaling ◽  
Ad Hoc ◽  
Solution Space ◽  
Real Data ◽  
Neural Network Modeling ◽  
Data Sets ◽  
Regularization Technique ◽  
Clustering Approach ◽  
The Relationship

The common neural network modeling practice of representing the elements of a task domain in terms of a set of features lacks justification if the features are derived through some form of ad hoc preabstraction. By examining a featural similarity model related to established multidimensional scaling techniques, a neural network is developed that generates features from similarity data and attaches weights to these features. The network performs a constrained search of a continuous solution space to determine the features and uses a previously developed regularization technique to minimize the number of features it derives. The network is demonstrated on artificial data, from which it recovers known features and weights, and on two real data sets involving the similarity of a set of geometric shapes and the abstract conceptual similarities of the 10 Arabic numerals. On the basis of these results, the relationship between the multidimensional scaling approach adopted by the network and an alternative additive clustering approach to feature extraction is discussed.

Evaluating Models of Attitude Structure

1972 ◽  
Vol 9 (3) ◽  
pp. 279-286 ◽  
Author(s):  
George S. Day
Keyword(s):  
Multidimensional Scaling ◽  
Building Material ◽  
Industrial Building ◽  
Cognitive Consistency ◽  
Consistency Theory ◽  
Affective Components ◽  
Relationship Of ◽  
The Relationship

Models of attitude structure differ primarily in the specification of the relationship of the cognitive and affective components. Two approaches to this relationship, based on cognitive consistency theory and multidimensional scaling techniques, are compared here. An application of the latter method to the analysis of a new industrial building material is presented.

Multidimensional Scaling and Application in Traffic Jam Prediction

2013 ◽  
Vol 291-294 ◽  
pp. 3060-3063 ◽  
Author(s):  
Bin Hui Wang ◽  
Lin Cai ◽  
Ming Liu
Keyword(s):  
Multidimensional Scaling ◽  
Travel Time ◽  
Statistical Methods ◽  
Travel Times ◽  
Traffic Jam ◽  
Traffic Conditions ◽  
Informational Technology ◽  
Wide Range ◽  
The Relationship

Many methods thus have been proposed to predict traffic conditions. However, it is difficult to accurately predict traffic jam, because it requires a wide range of knowledge such as statistics and informational technology. It is known that the probability of traffic jam can be evaluated by travel times of passing cars in a location of the motorway. In this paper, we restrict our attention to finding more efficient statistical methods through comparing models. For this reason, we used Multidimensional Scaling statistical methods to study the relationship between traffic conditions and travel time in different locations and times. This work aims at applying basic models to forecast traffic conditions.

scholarly journals Multidimensional Scaling in Market Research: Advantages and Disadvantages

2010 ◽  
Vol 49 (1) ◽  
pp. 52-61
Author(s):  
Angelė Kėdaitienė ◽  
Vytautas Kėdaitis
Keyword(s):  
Multidimensional Scaling ◽  
Graphical Models ◽  
Euclidean Distance ◽  
Interpoint Distance ◽  
Final Model ◽  
Advantages And Disadvantages ◽  
Similarity Judgments ◽  
Object Similarity ◽  
The Relationship

Multidimensional scaling was developed by psychometricians, namely R. N. Shepard (1962) and J. B. Kruskal (1964). Its purpose is to deduce indirectly the dimensions a respondent uses to evaluate alterna­tives. The reason for using the indirect approach is that, in many cases, the attributes may be unknown and respondents may be unable or unwilling to repre­sent their reasons accurately. As already mentioned, multidimensional scaling requires an object-by-object similarity matrix as an input. Initially popularized, however, multidimen­sional scaling relies on judged similarity. That is, re­spondents indicate how similar pairs of objects are directly rated (e.g. on a 1–10 scale). This can be a bur­densome task since for p objects p(p-1)/2judgments are needed. Still, the use of similarity judgments is relatively easy for respondents, especially when they cannot or do not want to reveal the basis for their opinion. The results of multidimensional scaling depend on (a) the sample chosen to judge similarity and (b) the objects whose similarity is judged and the quality of input data. Multidimensional scaling derives dimen­sions that appear to be used by those rating a par­ticular set of objects. The basic type of multidimensional scaling in­volves deducing graphical models of alternatives (e.g. brands) alone (simple space) from similarity data. Some early applications of multidimensional scaling accepted apparent dimensions as “truth” without question or validation, which often proved to be disastrous. It is advisable to use multidimensional scaling as a generator of hypotheses rather than as a final model of the market. Any important result should be confirmed on a separate sample with a separate method, such as direct questioning, before the results are given too much credence. Multidimensional scaling generates a configu­ration in which the relative positions of the brands are unique. The picture can be changed by several opera­tions without changing the relationship among the interpoint distance in some of the algorithms (as­suming the Euclidean distance is used, which it almost always is). A major problem in data collection is the bur­den on respondents as the number of alternatives increases (e.g. 20 alternatives require 190 pairs). However, if respondents are “homogeneous”, it is possible to have different subjects rate a different pair.

Political Perception Dimensions Based on Negative Similarities

1979 ◽  
Vol 49 (1) ◽  
pp. 19-26 ◽  
Author(s):  
Bo Ekehammar ◽  
Jim Sidanius
Keyword(s):  
High School ◽  
High School Students ◽  
Political Parties ◽  
Multidimensional Scaling ◽  
School Students ◽  
Scaling Solution ◽  
Political Perception ◽  
The Relationship

The purpose was to study (a) the relationship between positive and negative similarity estimations for political stimuli and (b) the generality of dimensions of political perception across methods and populations. One sample of 199 high-school students gave estimates of the degree of positive similarity between all pairs of 8 Swedish political parties on an 11-step scale. Another sample of 148 high-school students provided the same type of data with the exception that they were also allowed to express negative similarities when the stimuli were perceived as opposite to each other. Contrary to the results from a previous study, the relationship between positive and negative similarities was linear, and the similarity distributions did not differ as expected. Possible explanations for these discrepancies are discussed. In accordance with expectations, the multidimensional scaling solution based on negative similarities was more parsimonious. Dimensions obtained were the same as those in an earlier study, which supports the generalizability of these dimensions of political perception.

Sign in / Sign up

Export Citation Format

Share Document