1
vote
0answers
39 views

How does Gower distance work with free text?

The Gower distance measure is a good measure for mixed-type data (i.e., data attributes can be qualitative, categorical, ordinal or binary). But can data attributes be free-text (e.g., names of ...
57
votes
6answers
6k views

Why is Euclidean distance not a good metric in high dimensions?

I read that 'Euclidean distance is not a good distance in high dimensions'. I guess this statement has something to do with the curse of dimensionality, but what exactly? Besides, what is 'high ...
4
votes
2answers
113 views

Is there an advantage to squaring dissimilarities when using Ward clustering?

Is there a reason to prefer squaring or not squaring the dissimilarities when clustering with Ward's method? The question is motivated by the following statement in the documentation for R's ...
0
votes
1answer
38 views

One huge cluster + small ones with vector-space model + cosine distance

I'm trying to cluster meaningfully a set of objects characterized by a vector space (bag-of-words) model. Each of those 5000 objects has 1-8 features ("words") from a set of 5500 possible. I used a ...
0
votes
1answer
34 views

Silhouette scores for different distance metrics

I clustered a data set using PAM with a euclidean distance metric and a pearson correlation distance metric. The average silhouette value of the correlation clusters is higher at most points than the ...
2
votes
0answers
35 views

What is a good technique for grouping objects based on binary or dichotomous traits?

I have a set of objects each of which has a list of traits. Data on the traits is binary: an object has a trait or does not. The number of objects that I have is moderately greater than the number ...
0
votes
2answers
30 views

How to cluster users based on search terms

How to cluster based on what users are searching on I'm working on an app which includes search functionality: a search box that allows a user to enter text and search the entire site. I have access ...
1
vote
1answer
80 views

String clustering and centroid computation

I have a text file document containing a set of words strings that I want to cluster. I want to use the K-means algorithm. As a ...
1
vote
1answer
46 views

Calculating (dis)similarity between different types of features

Disclaimer: I understand that this question is specific to the types of data, the end goal, etc. but I just wanted to get some quick tips regarding calculating dissimilarity between different types of ...
0
votes
1answer
61 views

distortion function for k-means algorithm

I was reading Andrew Ng's ML lecture notes on K-mean clustering, in which the distortion function is defined as follow $$J(c,\mu) = \sum^m_{i=1} || x^{(i)} - \mu_{c^{(i)}}||^2$$ I am puzzled about ...
0
votes
3answers
2k views

Why does k -means clustering algorithm use only euclidean distance metric?

Is there a specific purpose in terms of efficiency, functionality why k-means algorithm do not use cosine similarity as a distance metric and use the euclidean norm
2
votes
2answers
169 views

The right distance for the clustering. Maybe Mahalanobis?

I have to do a cluster analysis and I'm asking which distance should I used. I know that 99% of the clustering are made using a euclidean distance, but I heard about the Mahalanobis distance and it ...
1
vote
1answer
92 views

Measure of similarity/distance of data points in geographic space

Given two points $p_1=(x_1,y_1,t_1)$ and $p_2=(x_2,y_2,t_2)$, where $x$ and $y$ refer to the geographic coordinates in the plane, and $t$ to some measured value. Two distance measures to evaluate the ...
0
votes
1answer
160 views

Representative point of a cluster with L1 distance

The representative point of a cluster or cluster center for the k-means algorithm is the component-wise mean of the points in its cluster. The mean is chosen because it helps to minimize the within ...
4
votes
2answers
140 views

What distance method to use in this scenario?

I have a 10 dimensional space which contain points that contain a 1 or 0 . example of two points : point1 : 1,1,1,0,0,0,1,1,0,1 point2 : 1,0,1,0,0,0,1,0,0,0 ...
0
votes
3answers
240 views

Is it possible to use Hellinger distance for environmental variables?

Here is the problem, Euclidean distance is not recommended for datasets with many zeroes (like matrices of species/site), as there is the risk of the abundance paradox (Orloci, 1978). Whereas to ...
3
votes
1answer
579 views

Measurement of similarity for hierarchical clustering trees

I am performing a number of hierarchical clusters on a dataframe of patient records (e.g. similar to http://www.biomedcentral.com/1471-2105/5/126/figure/F1?highres=y) I am experimenting with ...
3
votes
2answers
552 views

What are distances between variables making a covariance matrix?

I have a $n \times n$ covariance matrix and want to partition variables into $k$ clusters using hierarchical clustering (for example, to sort a covariance matrix). Is there a typical distance ...
1
vote
1answer
116 views

Distances vs. “distance like functions” in clustering

I am studying Kogan's "Introduction to Clustering Large and High Dimensional Data" because I would like to better understand clustering (I never worked with it). Until now "clustering" means to me to ...
4
votes
2answers
2k views

Numerical Instability of calculating inverse covariance matrix

I have a 65 samples of 21-dimensional data (pasted here) and I am constructing the covariance matrix from it. When computed in C++ I get the covariance matrix pasted here. And when computed in matlab ...
4
votes
1answer
340 views

What are the use cases related to cluster analysis of different distance metrics?

I'm trying to use different distance metrics like Euclidean, Manhattan, cosine, chebyshev among other distance metrics in my k-means algorithm to calculate distances between the data points and the ...
5
votes
2answers
368 views

Calculating similarity and clustering question

I have a dataset of about a million companies containing their names, total employees and annual sales. I want to come up with a function that when given the company returns the 5 most similar ...
3
votes
2answers
179 views

What function of distance for the questionnaire data?

I have data from questionnaire from school. First question is study program (only 2 programs) and next 35 questions are various questions (influence of friends etc.) Possible answers for 35 questions ...
0
votes
2answers
318 views

Interpret Silhouette plot for large microarray dataset

For a microarray experiment with ~40,000 probes and ~30 samples I used the clara function from R to cluster my expression matrix. How do I interpret this silhouette plot? Firstly, I don't ...
1
vote
3answers
219 views

Clustering algorithm and distance function for sets

I am willing to run a clustering algorithm on data records consisting in sets each one representing the features enabled at a certain time. Is there any clustering algorithm you would recommend me to ...
2
votes
2answers
350 views

(hierarchical) cluster analysis with non-standard distance

My question is triggered by a question that was asked on stackoverflow: http://stackoverflow.com/questions/12198115/using-different-metric-for-hclust-linkage. The thing is this: I can formulate an ...
3
votes
1answer
1k views

Euclidean Distance b/t unit vectors or cosine similarity where vectors are document vectors

I was reading Similarity Measures and suddenly my whole world was falling apart. I have implemented a search engine using clustering techniques. For clustering, I used k means which uses Euclidean ...
0
votes
3answers
2k views

Using a cosine similarity does not work for any dataset

I have a clustering algorithm, where if I use an euclidian distance as similarity, it works well on any dataset. If I replace it by a cosine similarity (see my code bellow), it will give a degenerate ...
0
votes
1answer
101 views

How can I replace this condition by a probability?

I want to see if a datapoint x should (or not) be assigned to a nearest component y using the following condition: if ($d > T$) then {do not assign x to y}. With $d = distance(x,y)$ and $T = ...
0
votes
1answer
232 views

Assigning new observations into existing clusters made from a distance matrix

I am building a classification model (binary outcome) and would like to include an external cluster membership code as a predictor. For training the model, this is straight forward. When "scoring" new ...
3
votes
2answers
1k views

How to input self-defined distance function in R?

I want to know how to to input a self-defined distance in R, in hierarchical clustering analysis. R implements only some default distance metrics, for example "Euclidean", "Manhattan" etc. Suppose I ...
1
vote
1answer
652 views

Entropy based on euclidian distances between datapoints / clusters centers?

Is it possible to define any useful entropy or conditional entropy which is based on the distance between datapoint(s) and cluster center(s), instead of basing on the number of points assigned to ...
4
votes
5answers
259 views

Significance of difference using a distance matrix

So my data consists of a distance matrix for some points, and a table classifying the points as either red or green. I want to know if there is any difference between the red and green points. I did ...
2
votes
2answers
868 views

How to derive a distance function based on multiple variables for cluster analysis?

I am not a statistician, so please excuse my lack of statistics knowledge/terminology. I have bunch of network nodes that I want to run cluster analysis on and identify clusters. So as far as I ...
6
votes
1answer
1k views

Hierarchical clustering with mixed type data - what distance/similarity to use?

In my dataset we have both continuous and naturally discrete variables. I want to know whether we can do hierarchical clustering using both type of variables. And if yes, what distance measure is ...
4
votes
2answers
778 views

Will the silhouette formula change depending on the distance metric?

I am using Silhouette width to compute the best value for k in k-means. As I am performing document clustering, I am calculating the values of a and ...
13
votes
3answers
400 views

$L_1$ or $L_.5$ metrics for clustering?

Does anyone use the $L_1$ or $L_.5$ metrics for clustering, rather than $L_2$ ? Aggarwal et al., On the surprising behavior of distance metrics in high dimensional space said (in 2001) that $L_1$ ...
10
votes
3answers
2k views

Is it ok to use Manhattan distance with Ward's inter-cluster linkage in hierarchical clustering?

I am using hierarchical clustering to analyse time series data. My code is implemented using the Mathematica function DirectAgglomerate[...], which generates ...
6
votes
1answer
468 views

Derivation of distance in two-step clustering

I am working with the two step cluster process in SPSS Modeler (Clementine) and trying to get a sense for the distance function used. It is a log-likelihood function (as stated in docs) but I am ...
8
votes
1answer
3k views

Clustering: Should I use the Jensen-Shannon Divergence or its square?

I am clustering probability distributions using the Affinity Propagation algorithm, and I plan to use Jensen-Shannon Divergence as my distance metric. Is it correct to use JSD itself as the distance, ...