文档下载 文档下载导航
设为首页 | 加入收藏
搜索 请输入内容:  
 导航当前位置: 文档下载 > 所有分类 > IT/计算机 > 互联网 > MDS多维标度
侵权投诉

MDS多维标度

ENCYCLOPEDIA

OF COGNITIVE SCIENCE

(to appear in 2002) ©Macmillan Reference Ltd

two or three dimensions, researchers can visually study the structure in the data. It also has been used to discover the mental representation of stimuli that explains how similarity judgments are generated. Sometimes, MDS reveals the psychological dimensions hidden in the data that can meaningfully describe the data. The multidimensional representations resulting from MDS are also often useful as the representational basis for various mathematical models of categorization, identification, and/or recognition memory (Nosofsky, 1992) or generalization (Shepard, 1987).

We will illustrate some of the issues in MDS with the analysis of a face similarity judgement task. In Figure 1a, the average of a group of subjects’ similarity ratings is shown for 10 faces shown in Figure 1c. The idea is to reveal some of the perceptual dimensions that subjects might have used when generating similarity judgments for these faces.

Techniques for MDS

There are many different MDS techniques to analyse proximity data and many issues in the analysis and interpretation of the results. First, there is the distinction between metric and nonmetric MDS. The goal of metric MDS is to find a configuration of points in some multidimensional space such that the interpoint distances are related to the experimentally obtained similarities by some transformation function (e.g., a linear transformation function). If the proximity data are generated with Euclidian distances for some stimulus configuration, then a procedure called classical metric MDS (Torgeson, 1965) can exactly recreate the configuration of points. Because a closed form solution exists to find such a configuration of points, classical metric MDS can be performed efficiently on large matrices. In nonmetric MDS (first devised by Shepard in 1962), the goal is to establish a monotonic relationship between interpoint distances and obtained similarities. The advantage of nonmetric MDS is that no assumptions need to be made about the underlying transformation function; the only assumption is that the data is measured at the ordinal level.

Kruskal (1964) proposed a measure for the deviation from monotonicity between the distances dijand the observed dissimilarities

Multidimensional Scaling

Mark Steyvers Stanford University

Assumptions of MDS

Multidimensional Scaling (MDS) describes a family of techniques for the analysis of proximity data on a set of stimuli to reveal the hidden structure underlying the data. The proximity data can come from similarity judgments, identification confusion matrices, grouping data, same-different errors or any other measure of pairwise similarity. The main assumption in MDS is that stimuli can be described by values along a set of dimensions that places these stimuli as points in a multidimensional space and that the similarity between stimuli is inversely related to the distances of the corresponding points in the multidimensional space. The Minkowski distance metric provides a general way to specify distance in a multidimensional space:

n

dij=

xik xjk

k=1

r

r

,

where n is the number of dimensions, and xik is the value of dimension k for stimulus i. With r =2, the metric equals the Euclidian distance metric while r=1 leads to the city-block metric. A Euclidian metric is appropriate when the stimuli are composed of integral or perceptually fused dimensions such as the dimensions of brightness and saturation for colours. The city-block metric is appropriate when the stimuli are composed of separable dimensions such as size and brightness (Attneave, 1950). In practice, the Euclidian distance metric is often used because of mathematical convenience in MDS procedures.

MDS can be applied with different purposes. One is exploratory data analysis; by placing objects as points in a low dimensional space, the observed complexity in the original data matrix can often be reduced while preserving the essential information in the data. By a representation of the pattern of proximities in

第1页

你可能喜欢

  • 应用多元统计分析
  • 厦门大学多元统计
  • 多维尺度分析
  • 数据挖掘

热门文档

相关文档

  • 多维标度法(MDS)的改进及其在国际市场研究中的应用_免...

    多维标度法(MDS)的改进及其在国际市场研究中的应用多维标度法(MDS)的改进及其在国际市场研究中的应用隐藏>> ? 1995-2004 Tsinghua Tongfang Optical Disc Co.,...

  • 第10章 多维标度法

    本章仅介绍常用的古典多维标度法和权重多维 标度法。 第二节 古典多维标度法 (Classical MDS) 一 相似与距离的概念 二 古典多维标度分析的思想及方法三 度量...

  • 多维尺度分析MDS

    23页 免费 MDS多维标度 5页 免费 市场分析方法LV11 多维尺... 暂无评价 9页 免费 多维尺度分析 14页 1下载券喜欢此文档的还喜欢...

  • 多维标度

    第十章 多维标度法第一节 第二节 第三节 第四节 引言 古典多维标度法(Classical 古典多维标度法 MDS) 权重多维标度(WMDS) 权重多维标度 实例分析与计算实现...

  • 多维标度

    多维标度_经济/市场_经管营销_专业资料。第十章第一节 引言 多维标度法 第二节 第三节第四节 古典多维标度法(Classical MDS) 权重多维标度(WMDS) 实例分析...

  • 第十章 多维标度法

    本章仅介绍常用的古典多维标度法和权重多维 标度法。 第二节 古典多维标度法 (Classical MDS) 一 相似与距离的概念 二 古典多维标度分析的思想及方法三 度量...

  • 多维标度法汇报学习

    多维标度法汇报学习_教育学/心理学_人文社科_专业资料。多维标度法 ? 第一节 第二节 第三节 引言 古典多维标度法(Classical MDS) 权重多维标度(WMDS) ? ...

  • 运用多维尺度标度法对NBA球员的聚类分析

    第二章 多维标度法简介多维标度法(multidimensional scaling MDS)是一种把高维降为低维,在低维空间展 示距离数据结构的多元数据分析技术,简称 MDS。多维标度法...

  • 多维标度法的聚类分析_问题与解法

    多维标度法的聚类分析_问题与解法_工学_高等教育_教育专区 暂无评价0人阅读0次下载举报文档 多维标度法的聚类分析_问题与解法_工学_高等教育_教育专区。MDS ...

  • 基于RSSI的多维标度室内定位算法_石欣

    满足非度量多 维标度法对相异性数据的要求。因此本文提出的RSSICMDS 算法, 直接使用 RSSI 值进行定位。 基于 RSSI 构建多维标度法中的相异性矩阵 P , 计算出...

站点地图 | 文档上传 | 侵权投诉 | 手机版
本站所有资源均来自互联网,本站只负责收集和整理,均不承担任何法律责任,如有侵权等其它行为请联系我们.
文档下载 Copyright 2013 doc.guandang.net All Rights Reserved.  -  email