Technique that renders observed or computed (dis)similarities among objects into distances in a low-dimensional space (usually Euclidean). It thus constructs dimensions for the data; the objects can be plotted and conceptualized in those dimensions

learn more… | top users | synonyms (1)

2
votes
0answers
41 views

Non-metric Stress in 3 way Multidimensional Scaling (INDSCAL): individuals vs. group

My data consists of 19 participants giving dissimilarity ratings for every possible pairwise comparisons of a set of 12 epistemic adverbs (so each participant gives 66 ratings). The goal is to ...
0
votes
0answers
18 views

How proximity and multidimensional scaling relate? #random forest

I have downloaded randomForest package of Breiman and try to use function MDsplot to plot the proximity of the data like the example in the manual ...
0
votes
0answers
27 views

Can I run a multidimensional scaling analysis with purely categorical data?

I have data where participants categorized facial expressions using one of seven emotion labels (Angry, Disgusted, Fearful, Happy, Neutral, Sad, and Surprised). Can I take the resulting confusion ...
0
votes
1answer
30 views

How do I apply MDS analysis on my data set?

Consider the following dataset (it is the emission probability matrix of a Hidden Markov Model): ...
2
votes
2answers
71 views

Understanding differences between large and small dimensional data when implementing algorithms

I'm working on a local outlier factor implementation based on the wikipedia entry : http://en.wikipedia.org/wiki/Local_outlier_factor This article seems to explain it in just two dimensional data. ...
0
votes
0answers
10 views

Add species characteristics, alongside count data, when using ordination

I have count data by site for 15 species, however I also have average size by species by site. I'm looking to use ordination techniques (non-metric multidimensional scaling) to reduce dimensionality ...
0
votes
0answers
58 views

Multidimensional Scaling: Interpreting output of different distance matrices (Euclidean or correlation)

I would like to understand the difference between using a Euclidean distance matrix or correlation matrix as input to a nMDS algorithm. I have completed MDS plots of both, and while similar, the ...
3
votes
1answer
91 views

Reference for dimension reduction techniques

This is a follow-up question to Is PCA appropriate for comparing subsets of panel data?. It turns out that, yes, PCA is appropriate. But there are also many other ways to reduce n-dimensional data to ...
1
vote
0answers
102 views

nMDS in vegan for soil data

I am working with abiotic soil data such as bulk density, moisture levels and soil chemistry as response data (some quantitative some as percentages) and a mix of abiotic and biotic data as ...
2
votes
0answers
28 views

Using the results of a non-metric multidimensional scaling to derive a measure of deviance

I'm working with a dataset containing the taxonomic compositions of floras all over the world and performed a non-metric multidimensional scaling (NMDS) based on the relative proportions of plant ...
0
votes
0answers
22 views

Multidimensional Scaling - external scale regression using standardized or unstandardized weights

I am having trouble determining whether to use standardized or raw regression weights when using multiple regression to interpret Multidimensional Scaling plots. Different textbooks seem to advocate ...
1
vote
1answer
84 views

Normalizing Vs. Scaling

Are the concepts of normalizing and scaling of data in conflict with each other? I am adding weights to my features, I have tried normalizing the weights and it didn't make any difference in the ...
0
votes
1answer
18 views

should weights be scaled too?

I am using supervised learning algorithms (specificly SVM) on my data. I know that scaling was needed for my input data. however as I am also adding weights (using pairwise comparison), I am not sure ...
0
votes
1answer
88 views

How to : a brief intro to scaling and rescaling data ( inputs) for supervised learning algorithms

I understand the concept of scaling and that it improves results in SVM's and NN's. however I would like to find somewhere where is is explained, in easy "layman's terms" terms. of how it is done. I ...
1
vote
1answer
40 views

Scaling in SVM (why and how to , plus references)

Hi I know why feature scaling is preferred in SVM, I have two questions: 1-does anyone know of legit articles of books explaining it. I am writing my thesis and I need references. It doesnt have to be ...
0
votes
0answers
45 views

Is subtracting the mean from PCA necessary when using an SVD result that is feature scaled?

I've applied SVD to the original data matrix and eliminated insignificant columns and rows from U and V^T respectively using the Sigma values. I multiplied together my optimized U, Sigma, and V^T ...
1
vote
1answer
53 views

Is it posible to perform the inverse of multidimensional scaling analysis

We have lot of 3D data and we reduced it to 2D for performing fuzzy clustering and obtaining prototypes. We used some matlab functions that were very well documented. Now we would like to see to which ...
4
votes
1answer
205 views

How to distance and to MDS-plot objects according their complex shape

Suppose I have four basal forms of signal (blue, purple, red, green). I also have created transition forms between each other. If you carefully look on the picture below, you can see that for example ...
1
vote
0answers
25 views

Suitable plot for 5 dimensional feature vectors?

I have a list of personality scores obtained from 100 people, based on the Big-Five personality test. Each person has one score for each of the five assessed traits. I put these scores into a 5 ...
0
votes
0answers
42 views

Converting data from classical scaling in MDS

I implemented the classical scaling and obtained the relative coordinates in Java. I want to translate these relative positions, but I am not getting how I should be doing this. This is just like: ...
0
votes
0answers
68 views

Is it correct to combine PCA and NMDS axes in a multiple regression?

I am considering to do a multiple regression in which some of the predictive variables are PCA (principal components) axes whereas others are NMDS (nonmetric muliple dimension scaling) axes. I would ...
2
votes
1answer
123 views

Multidimensional Scaling “eurodist”

I have a question regarding Multidimensional Scaling. I used the dataset eurodist from the package datasets to generate a 2 ...
0
votes
0answers
118 views

Draw biplot on nonclassical MDS interpration?

How can I draw similar biplot like you can on PCA, so I could compare these two visualization techniques / methods? I got dissimilarities and distances of my data in simple plot, but I would like it ...
0
votes
1answer
439 views

Multidimensional scaling using Python

I have 6,000 points for which I have all pairwise distances in a distance matrix. I want to get an idea whether these data were generated by a mixture of Gaussian distributions so I'm trying to get a ...
0
votes
0answers
179 views

What exactly is the vector fitting on a nMDS plot telling me about community structure?

I think I have a basic misunderstanding of what the vectors typically plotted over an nMDS (non Metric Multidimensional Scaling) plot are telling me. I understand the direction of the vectors point to ...
0
votes
0answers
43 views

How to de-trend the data before doing MDS?

I am given a set of regions' macro-economical development indices, such as number of unemployed, gross value added, and so on. Each index naturally is correlated with a size of the region. I want to ...
1
vote
0answers
32 views

Multidimensional scaling of variables with multiple sub-features?

Let's say I have a year's worth of magazine issues (January, February, March, etc), and I want to visualize the differences among them. The classic example of multidimensional scaling (MDS) would have ...
0
votes
0answers
52 views

Scaling data in machine learning algorithm

I am currently working on a linear classifier, which uses the statistical learning paradigm; that is, no knowledge about the distribution from which that training data are drawn from is available. ...
1
vote
0answers
48 views

Steps to follow for correspondence analysis when each brand is not shown to every respondent

I want to understand the steps followed for correspondence analysis when each brand is not shown to every respondent. Till now I used to assign a number (proportion) to each brand for each attribute ...
1
vote
1answer
415 views

Guttman's smallest space analysis

I am trying to replicate an analysis as conducted by Raven, et al. (1998). A scale is analyzed with Guttman's smallest space analysis. First of all I wasn't able to find a suitable procedure in R and ...
1
vote
1answer
143 views

Kernel PCA with an SVD algo

Suppose that I have a great algo for calculating the SVD and I want to do Kernel PCA. It is possible to first apply the Kernel function to my data and then run the SVD algo on the transformed data?
1
vote
1answer
69 views

Multidimensional quantiles

I have 1000 observations with 2 continuous variables : Observation ID | X | Y
1
vote
0answers
202 views

Gower distance and MDS: How to determine which variables count?

I have morphological data from two different determined groups (It and Nd), where the variables are heterogeneous (continuous, ...
2
votes
0answers
100 views

When using Nonmetric Multidimensional Scaling, is there an explanatory metric similar to loadings in PCA?

As a beginner to MDS, here is my thought process: Given a data set of environmental factors that may effect a certain sites, when I run a PCA on each site I get a list of principal components. If I ...
0
votes
0answers
393 views

Multidimensional scaling in R with Spearman's rank correlation

I'm new to MDS, but I found some good starter code here (http://mhermans.net/static/postdata/r-examples/neighbours-mds/neighbours-mds-example.html) and which I borrowed for below. The code works fine, ...
6
votes
2answers
860 views

Perform feature normalization before or within model validation?

A common good practice in Machine Learning is to do feature normalization or data standardization of the predictor variables, that's it, center the data substracting the mean and normalize it dividing ...
0
votes
1answer
88 views

Multidimensional scaling for big dissimilarity matrix

I have a large symmetrical dissimilarity matrix of dimension 300 000. Can you please suggest the multidimensional scaling algorithms that can work with such large data? Input of course can be the ...
1
vote
1answer
59 views

Which values to use for scaling out of sample PCA data

I have centered and scaled inputs via prcomp (): prOut<-prcomp(trainSet[,2:4],scale = TRUE,scores=TRUE) I now want to use ...
0
votes
0answers
58 views

Transforming distance matrix to univariate data

Is there a valid mathematical or statistical approach to converting site by site distance objects/matrices to a univariate site vector or column in a dataframe? I have a Euclidean distance matrix for ...
2
votes
0answers
163 views

Why does some model-based clustering fail to fit with a large number of dimensions?

I am attempting to cluster data using Mclust. The data is originally from a dissimilarity matrix, transformed via multidimensional scaling in R (MASS::isoMDS). As I ...
3
votes
1answer
219 views

whether to rescale indicator / binary / dummy predictors for LASSO

For the LASSO (and other model selecting procedures) it is crucial to rescale the predictors. The general recommendation I follow is simply to use a 0 mean, 1 standard deviation normalization for ...
0
votes
1answer
231 views

Analysing data measured as proportional composition

I have a data set on the proportional composition of marine substrate for different locations which I would like to compare. For example, one replicate transect within a location may be 50% sand, 25% ...
2
votes
0answers
492 views

How to interpret variation explained by principal coordinates?

I have recently seen a couple of Principal Coordinates Analysis (PCoA) projection plots which show "percentage variation explained" by the respective principal coordinates. Given that the analysis is ...
1
vote
0answers
24 views

Issue of multiple scaling method

I plan to use multiple scaling in R and start with a toy example. There are two matrices. The first contains 10 observations and the second one contains two replicated rows. It appears that the ...
2
votes
0answers
24 views

How to enhance the performance of naive CMDS?

I'm performing Classical MDS on the distance over columns of binary matrix. The result is like this: The points lies on two lines vertical to each other. I don't think useful information is shown ...
8
votes
3answers
360 views

How to project high dimensional space into a two-dimensional plane?

I have a set of data points in a N-dimensional space. In addition, I also have a centroid in this same N-dimensional space. Are there any approaches that can allow me to project these data points into ...
3
votes
1answer
328 views

MDS and PCA eigenvalues and eigenvectors

I understand that Multidimensional scaling (MDS) is same as doing Principal Components analysis (PCA) if Euclidean distance is used, this is known as Metric MDS. But I came across this in a book that ...
0
votes
0answers
72 views

Multidimensional scaling

I am running into some problems performing a multidimensional scaling image. First of my dataset is quite large (330.000 fields: 33000 rows, 10 columns). The output image needs to contain 10 dots, 1 ...
0
votes
1answer
143 views

NMDS: why is the r-squared for a factor variable so low

I am doing an NMDS ordination. The data come from a number of sites scattered around two lakes. In the plots, I coloured the samples from the two lakes blue and green. There seems to be some pretty ...
2
votes
1answer
619 views

NMDS and variance explained by vector fitting

I just ran a non metric multidimensional scaling model (nmds) which compared multiple locations based on benthic invertebrate species composition. After running the analysis, I used the vector fitting ...