Partitioning data into subsets of objects according to their mutual "similarity," without using preexisting knowledge such as class labels. Clustered-standard-errors and/or cluster-samples should be tagged as such; do not use the "clustering" tag for them.

learn more… | top users | synonyms (1)

0
votes
0answers
13 views

Optimal grouping in one dimensional data with constraints [duplicate]

I have a 1d series of data with of approximately 100 values. I would like to partition series into 1, 2 or 3 groups, depending on the proximity of the values. What statistical technique would work ...
0
votes
1answer
14 views

Evaluating a clustering against multilabels

I have a clustering of text documents, where each document is uniquely assigned to a cluster. I have a set of labels (keywords) attached to each document. That is, each label may be applied to many ...
0
votes
0answers
16 views

Clustered Data and Kruskal–Wallis

I have a data set of loan amounts that is naturally clustered into 3 groups: 0 - 5 year loan 10 - 15 year loan 20 - 30 year loan The 3 groups are not normally distributed. Since I don't need to ...
3
votes
1answer
35 views

Under what conditions would clustering on top of Principal Components would return different result (and worse) than clustering on the data itself?

Since Principal components capture most of the information, clustering on them should provide similar result as that of the clustering on the original data. As such, it seems to me (who's not a ...
0
votes
0answers
27 views

When is preferred the relative and stability-based cluster validation?

I need to validate a clustering algorithm result. I know that Cluster Validation is commonly divided into four categories: internal, external, relative and Stability-based criteria, where internal and ...
0
votes
0answers
28 views

“Pairwise dependence probability”

From pg 9 of [1]: The notion of dependence between two variables A and B is taken to be mutual information; the amount of evidence for dependence is then the probability that the mutual ...
0
votes
0answers
44 views

R: finding spatial patterns in rasters (Moran's I etc)

i'm not really experienced in spatial stats yet, but i'm growing into it. I basically want to ascertain if certain values in a raster are a) autocorrelated and b) are more likely to exist in a ...
-1
votes
1answer
27 views

Alternative of canopy clustering algorithm in K-means algorithm

I am analyzing implementation of K-means clustering algorithm in MadLib project. Here K-means algorithm uses Canopy clustering for initial set of Centroid.I am just wondering , are there any other ...
0
votes
1answer
49 views

Why is it bad to use Pearson distance in K-means clustering? [duplicate]

I have implemented this algorithm in MATLAB and when I produce plots I notice that using Euclidean distance, I usually get presented with a clear pattern (sum of squares decreases with the number of ...
0
votes
0answers
5 views

paired t-tests in cluster randomised trials

Let's say we have a cluster randomised trial which includes a within-cluster comparison of a normally distributed continuous variable (Y) from a control and treatment group. Let's say we take a ...
0
votes
1answer
53 views

Is that correct about dimensionality reduction and clustering?

Could you please help me to understand it because I'm not sure if I got it correctly. Let's say I have a dataset, of persons, with 100 features, various characteristics like height, weight, age, etc. ...
1
vote
0answers
39 views

Using entropy to imputing missing value based on grey relational analysis and clustering

This algorithm contain three techniques : 1-fuzzy c-mean clustering 2-Grey relational theory 3-Entropy multiple imputation The frame work of this algorithm is as follows : My questions are ...
0
votes
0answers
9 views

Investigate clusters of outcome variables in a study and verify whether stratification for an exposure changes the clusters

We have a study in which we have a number of outcomes (~30) and big number of observations/patients (> 1000) and we tested the effect of a certain exposure on the probability of these outcomes. For ...
0
votes
0answers
12 views

Confidence Score for Unsupervised randomForest?

I am using the unsupervised form of randomForest (R) as part of a clustering scheme for 1000 peaks based on 30 features. I use the proximity matrix produced by ...
0
votes
0answers
14 views

Clustering + SVM -> transductive SVM?

Given a binarily labeled train set, and an unlabeled test set, consider the following two-step classification system: step 1: the train and test data is clustered. step 2: an SVM is fitted for each ...
0
votes
0answers
38 views

Clustering of Line Graphs

I have a sample of line graphs (like the ones below), and I am trying to find the easiest way to cluster graphs with similar patterns. It would be great to be able to say something like "X% of graphs ...
1
vote
0answers
42 views

How to calculate the distance between clusters using log-likelihood in two step clustering? [closed]

Log-Likelihood Distance (TwoStep clustering algorithms) How to calculate the distance between clusters(i.e record and cluster) using log-likelihood in two step clustering using the below mentioned ...
0
votes
1answer
36 views

Weird Clusplot when plotting k-mediods clustering vector

The basic idea of the problem is that I need to cluster a set of points for which I have a dissimilarity matrix. I have a dataset of around 4600 points (latitudes and longitudes). I have also ...
0
votes
1answer
37 views

PCA vs. Spectral Clustering with Linear Kernel

Consider a feature vector matrix $X := [x_1 x_2 \dots x_d] \in \mathbb {R}^{n\times d} $ that I hope to use as part of some supervised learning procedure, say, regression. Suppose that also, $d \gg n ...
1
vote
0answers
29 views

Summarizing multiple clustering results

I'm working on a problem where observations are being clustered within groups but I'd also like to compare the groups. However I am not sure of the best way to compare the groups. In total I have ...
0
votes
1answer
36 views

Combine two, three, (n) metrics for calculating dissimilarity matrix

I have a data set with 9000 instances and 40 attributes of mixed data, that is categorical and numeric. My target is to group them into clusters using whichever clustering algorithm works best. I've ...
2
votes
1answer
34 views

How to decide the numbers of row & column clusters for co-clustering

I want to use the blockcluster package in R to perform co-clustering on my data. But the function requires the number of row and ...
0
votes
0answers
11 views

using histogram distances for clustering

Business context: for our retail client there are a few crucial products. All of them have 3 phases (introduction, stable phase and end of life). After correcting for seasonality and discounts (as ...
0
votes
1answer
32 views

cforest - Prediction without labels

I have some data that I want to get the important variables from. I want to use random forest to get this information. The problem is that the data does not have labels. From what I understand random ...
0
votes
0answers
14 views

Encoding a set of vectors into 1 vector

I have a set $X$ that contains a set of vectors $V_i$ and I want to create a vector $u$ that represents each set. $V_i \in X$, $u$ is a representation of $V_i$. What are the possible ways of doing ...
2
votes
2answers
53 views

Validate Cluster Analysis by doing it on two subsamples

I am working on validating a cluster analysis. I have read somewhere the approach to cross-validate the cluster analysis. The link of the article is ...
0
votes
0answers
35 views

Mathematical steps involved in calculating the gap statistic

Can anyone tell me the mathematical steps involved in calculating gap statistic? (That is, without using any external software.) This is my dataset, the output of my cluster analisys: ...
0
votes
1answer
22 views

Measure influence of attribues on clustering

I don't have a specific example for my problem and maybe this is trivial, but I want to know how to measure the influence of specific attributes (or dimensions) of a dataset for clustering, like there ...
1
vote
1answer
22 views

In R package 'cclust' is there an equivalent of 'nstart' option from the 'kmeans' package? [closed]

I am trying to do k-means clustering in R using the cclust package. In k-means clustering, the initial centroid assignment greatly affects the final allocation. The kmeans package has an nstart ...
0
votes
0answers
19 views

Model-based clustering methods for mixed data types

Are there any model-based clustering algorithms for mixed type data (i.e., with both categorical and continuous variables)? The popular 'mclust' can only handle continuous variables.
0
votes
0answers
29 views

How does neuralgas clustering work? Why is it better than k-means for handwritten digits?

Recently I tried to cluster the semeion.data file available in UCI repository using kmeans clustering as well as using the ...
0
votes
0answers
12 views

Finding the stability of selected parameter values?

I have a system (not a predictive model) that will produce four results (R1 to R4) given a set of input data. The system can be tuned using four parameters (P1 to P4). I can find the maximum value for ...
0
votes
2answers
23 views

Clustering of binary/nominal variables in one sample

Assume that a medical school classifies its active, full-time students according to their free time activities. By distributing simple questionnaires individuals have to answer whether or not they're ...
1
vote
2answers
67 views

Clustering with categorical and numeric data [duplicate]

I frequently come across data sets that have both categorical and numeric data. I think this is just a fact of life where the data is not all in one category. I'm basically trying to find some ...
1
vote
0answers
14 views

A situation where ignoring clustering optimises the Type I and Type II error rates?

I am interested in modelling clustered data with a small number of clusters as follows: $$Y_{ij} = β_0 + β_1X + u_i + e_{ij}$$ (where $_i$ = 1 to 3; $_j$ = 1 to 12); $Y_{ij}$ is our normally ...
1
vote
1answer
32 views

Segmentation analysis based on Likert [closed]

I would like to do a segmentation analysis based on a Likert scale survey. I want to ask respondents to rate certain attributes on a scale of 5 points (not valuable at all, not very valuable, neutral, ...
0
votes
0answers
12 views

feature selection for clustering: wrappers

I am trying to understand if it is correct to perform a feature selection process using a wrapper method (for example using algorithms such as random forest, linear regression etc.) and then to extend ...
1
vote
1answer
45 views

Markov Cluster Algorithm transition matrix

I am reading the notes on Markov Cluster Algorithm by Kathy Macropol (http://www.cs.ucsb.edu/~xyan/classes/CS595D-2009winter/MCL_Presentation2.pdf) On slide 14/46 the author talks about inflation and ...
0
votes
1answer
27 views

A problem on clustering with weakly-supervised data

I have a clustering problem, which type I can't define. It's not a pure unsupervised or semi-supervised learning. Training data consists of points, which belong to some clusters. I know, which points ...
1
vote
1answer
46 views

how to learn / imitate clustering algorithm?

I have a device which every 100 milliseconds receive 15 data-points, and return the same data-points divided into distinct subsets (each of size <= 4), using some unknown logic/method. My end goal ...
0
votes
1answer
36 views

Order a correlogram

I have 30 days data, and I am plotting correlogram using corrplot (R package). I have an option to order the correlogram with hclust option and hence I obtain plot as shown on right side of below ...
1
vote
2answers
60 views

Using clustering for unsupervised classification (visualizing k-means cluster centers)

I know that the cluster centroid is the middle of a cluster. It's a vector containing one number for each variable, where each number is the mean of a variable for the observations in that cluster. I ...
4
votes
0answers
50 views

How do i compare two Self Organizing Maps?

I have results (weights) for multiple runs of self organizing map. I am trying to compare these results to check if my algorithm gets to the same solution from different random initial weights. I have ...
0
votes
1answer
45 views

Alternate distance metrics for two time series

I have time-series data of different houses. Assume it is power consumption data. Now, I want to cluster the houses following similar power consumption pattern utmost. So, the various distance metrics ...
0
votes
0answers
12 views

Compact Spatial Clustering with exogen variable

I have a table with 2400 polygons being a partition of a country at a low level. I have as well the population and the area for each polygon, giving me the density of each polygon. I would like to ...
0
votes
0answers
15 views

Determining dates of regime switches in historical time series

I am interested in determining the dates of regime switches (i.e. starting points of segments of a (univariate) time series that show similar behaviour, such as similar volatility) in a historical ...
1
vote
0answers
25 views

Identify subgroup similar to another group [closed]

I have 2 groups of people (group2 is much larger than group1). The two groups have these attributes: demographic information and ...
-1
votes
2answers
41 views

Meaning of this Cluster Analysis

I have 801 households (or customers). I have say 100 features on which I will describe a customer. I have a feature map with me. I now apply K Means algorithm for the value of K say 6. I get 6 ...
0
votes
1answer
22 views

Why is K-Means++ SLOWER than random initialization K-Means?

K-Means is an iterative clustering method which randomly assigns initial centroids and shifts them to minimize the sum of squares. One problem is that, because the centroids are initially random, a ...
1
vote
1answer
22 views

How to interpret the PAM output

I am using the PAM function in R, and I don't understand how to evaluate its output. Whereas in K-means the ratio between the between sum of squares to the total sum of squares already gives a very ...