k-means is a family of cluster analysis methods in which you specify the number of clusters you expect. This is as opposed to hierarchical cluster analysis methods.

learn more… | top users | synonyms

1
vote
1answer
8 views

Need a little help understanding K-means++ seeding

I have been working on a project that involves using K-means clustering for generating adaptive palettes from images. I understand the general process of K-means clustering, and I understand the ...
1
vote
0answers
14 views

Controling segmentation process in order to get usable segments

My aim is to create segments based on survey data. This in it self is quite straight forward: I use PCA to extract information from the survey answers, and then ...
0
votes
0answers
12 views

Difference between criterion of k-means?

I am watching a video on k-means clustering here https://www.youtube.com/watch?v=sLf0Z9tCTjE&index=30&list=PL3DFCC23FCE3C7EFB, in which (12:14) the professor briefly mentioned some criterion ...
2
votes
0answers
27 views

Gaussian Mixture and K-Means ?! a big challenge?

This is taken from Tom. Mitche Material as Old-Exam. I think the (2) is true and not (3). Who can verify me?
-1
votes
0answers
19 views

how to divide my categorical data into three clusters [on hold]

I have some data of 10 attributes, all of which are categorical attributes (factors in R). I would like to divide this data into three homogeneous sets(clusters). I went about doing 3-means ...
0
votes
0answers
8 views

calculating distance among unordered set partitions for k-mean clustering?

I have a dataset for which I construct unordered set partitions for each data point, e.g. {{1,2,3}{4,6}{5}} for one and {{1,3}{2,4,5}{6}} for the next. I would like to perform k-means clustering on ...
0
votes
1answer
24 views

Chosing optimal k and optimal distance-metric for k-means [duplicate]

I have a data-set with roughly 20-dimensions and millions of points which I want to cluster. The goal is to find a set of clusters which: Are as distinct as possible from each other (minimum ...
2
votes
0answers
13 views

K-Means Voronoi Issue: Size?

http://en.wikipedia.org/wiki/Cluster_analysis It states that: K-means separates data into Voronoi-cells, which assumes equal-sized clusters (not adequate here) and ...
0
votes
0answers
9 views

Stata K-means Classification [closed]

I would like to know how to clasify a database in Stata using k - means. Thanks! Guillermo
0
votes
0answers
13 views

How to show output for K-means clustering on multi-dimensional data?

I have to implement K-means algorithm for K=10 on handwritten digits data. The data matrix is 2500 X 784,i.e there are 2500 data points each with dimensions 784 .After clustering,I have to label each ...
0
votes
1answer
43 views

What is the benefit of using Manhattan distance for K-medoid than using Euclidean distance?

Please give me the reasons. I didn't find any k-medoid example that's calculation is done using Euclidean distance. All examples are made of Manhattan distance for k-medoid.
2
votes
1answer
65 views

K-means for non-spherical (non-globular) clusters

It is said that K-means clustering "does not work well with non-globular clusters." However, is this a hard-and-fast rule - or is it that it does not often work? I have a 2-d data set (specifically ...
3
votes
7answers
281 views

Comparing k-means results to original data: how to interpret the resulting plots?

I'm running k-means on my dataset that can be found here that has 7 classes. I plotted the ggpairs for the dataset and then took k-means and plotted ggpairs again ...
1
vote
1answer
25 views

Why do final cluster centers change after applying results from past K-Means clustering (SPSS)?

I have a question regarding what happens after I apply k-means clustering centers to a new data set. Basically, I ran k-means clustering on a dataset1, saved the cluster centers, and applied it to a ...
2
votes
1answer
26 views

Statistical Classification Method to Compare PURELY Categorical Data?

I have about a half-dozen variables, each of which can have anywhere from three to ten outcomes. I have to measure the degree of separation/similarity between rows. Either we can do some sort of ...
-1
votes
1answer
31 views

convert categorical data into numerical data? [closed]

I am doing project for post graduation....project is document clustering.in the project raise the problem is the text data convert into the numerical data?
0
votes
1answer
33 views

Probability for selecting centroids - K-means++

K-means++ selects centroids one by one, where each point has the chance to become next centroid with probability proportional to distance to closest centroid already selected. I implemented it like ...
2
votes
1answer
31 views

Cluster analysis without knowing the structure of the data set

I’m working on a task regarding cluster analysis for about half a year now, but since the fields of pattern recognition and cluster analysis are quite complex ones, I would call myself a beginner in ...
1
vote
0answers
51 views

Gaussian mixture models restriction? [closed]

I read this note that with striction on GMM with some condition this algorithm is more like to K-means: the adaptations of the Gaussian mixture models algorithms with Restrict each $\Sigma_i$ ...
0
votes
0answers
17 views

Clustering not producing even clusters

I'm using k-means clustering to processes running on machines. Dataset sample : machine name, process m1,java m2,tomcat m1,word m3,excel Build a matrix of ...
2
votes
0answers
39 views

jenks natural breaks vs k-means

I am new to this topic. As far as I know both are data clustering methods. Then my question is when is Jenks prefered over k-means? I read on this website that jenks is particularly suited for ...
-1
votes
2answers
26 views

k-means nstart equivalent for EM Clustering? Report only the best solution from a large number of initializations?

In K-means clustering, you can specify an nstart=i parameter, which performs the algorithm i times (i.e. selects the initial k random centroids i times) sand reports the best answer only. If I perform ...
0
votes
1answer
50 views

K-mean clustering with unknown k

How do I perform k-mean clustering with unknown k? I also need to provide a confidence interval for k. I am thinking in the line of putting a Poisson prior on k. Does that make sense? Does there exist ...
0
votes
1answer
23 views

K-Medoids swapping inside clusters

I'm a bit confused with concept of K-medoids. It seems that original algorithm (PAM) describes that swap step should be performed by swaping only one of the medoids with one non-medoid point from ...
0
votes
1answer
28 views

K-means++ like initialization for K-medoids

Does it make sense to use initialization in K-medoids like in the case of K-means++? To be precise - is it good to select "farthest" points as initial medoids? (farthest in sense that points that are ...
3
votes
1answer
70 views

Understanding cluster plot and component variability

I have run k-means clustering. I have also plotted the results using the following code in R: ...
14
votes
2answers
412 views

Why does gap statistic for k-means suggest one cluster, even though there are obviously two of them?

I am using K-means to cluster my data and was looking for a way to suggest an "optimal" cluster number. Gap statistics seems to be a common way to find a good cluster number. For some reason it ...
0
votes
0answers
17 views

Clustering objects with missing values

I have some time-series that I would like to cluster, but they can have missing values. One approach that may be ad hoc is to use an algorithm like K-medoids, and to use similarity measure that will ...
0
votes
0answers
19 views

I understand the algorithm for k means.But don't understand on how to apply it on testing data

I am having problems not on understanding the k means algorithm but on how to apply it on training ,validation and testing data.Is it like this: Training phase: Apply k-means on the input data and ...
0
votes
1answer
22 views

Proof that points change clusters less often as iterations proceed in k means

Is there a way that to prove the following: In k-means clustering, as the iterations proceed, the data points tend to stay in their existing clusters, overall, because the replacement of the centroid ...
0
votes
1answer
188 views

Plotting k-means clustering with mixed numerical/categorical data

I have a dataset in CSV format that looks as follows: ...
2
votes
2answers
112 views

Clustering a correlation matrix

I have a correlation matrix which states how every item is correlated to the other item. Hence for a N items, I already have a N*N correlation matrix. Using this correlation matrix how do I cluster ...
0
votes
1answer
77 views

LDA, PCA and k-means: how are they related?

I am trying to understand how linear discriminant analysis (LDA) is related to principal component analysis (PCA) and k-means clustering method. As an example, here is a comparison between PCA and ...
1
vote
1answer
76 views

K-means initial centers membership

I'm trying to plot all the steps of a k-means algorithm with r, but I can't. The k-means algorithm works in this way: Step 1. Initialize the center of the clusters Step 2. Assign the closest ...
1
vote
2answers
110 views

Clustering based on large Jensen-Shannon Divergence distance matrix

I have a dataset with large number of features and about 15 000 observations. I’m using a probability distribution distance metric related to Jensen-Shannon divergence (JSD) to cluster the ...
6
votes
1answer
197 views

k-means|| a.k.a. Scalable K-Means++

Bahman Bahmani et al. introduced k-means||, which is a faster version of k-means++. This algorithm is taken from page 4 of their paper, Bahmani, B., Moseley, B., Vattani, A., Kumar, R., & ...
1
vote
2answers
111 views

K-means++ algorithm

I try to implement k-means++, but I'm not sure how it works. I have the following dataset: (7,1), (3,4), (1,5), (5,8), (1,3), (7,8), (8,2), (5,9), (8,0) From the wikipedia: Step 1: Choose one ...
1
vote
1answer
62 views
0
votes
0answers
36 views

Determining k in k-means clustering by community detection in graph

I am faced with a problem of choosing an appropriate number of clusters in highly dimensional data. I've read many approaches to determine the number of clusters, and finally came to a solution and I ...
0
votes
1answer
112 views

Outlier detection with data (which has categorical and numeric variables) with R

Scenario I have a project about fraud detection where i need to find outliers by kmeans. I have a dataset about bank credits length of 1000. There are 21 columns (14 categorical, 7 numeric ...
2
votes
1answer
69 views

K-means clustering feature selection

I have a set of English and foreign language documents that I would to perform k-means clustering on to find document groups by topic. These documents are concatenated social media comments for ...
2
votes
1answer
29 views

Can I use HCA-Ward's cluster-centers to run a K-means including a new item, to see to which cluster is more similar to?

Thank you for reading my question. I have an archaeological case-study, that we can call "Site1", that I want to compare with 9 others "Sites" studied by other scholars. For all of them I have 8 ...
-1
votes
2answers
46 views

Which K-mean algorithm I have to use for this problem?

Perform a k-means Clustering (non-iterative algorithm) using k=2 randomly initialised centroids (cluster prototypes), and the Euclidean distance. At the moment I manage to understand you can use ...
0
votes
1answer
25 views

Comparing / quantifying clusters

I have 'n' observations which are classified in two classes: Class A and Class B. The observations are mis-balanced with Class A constituting around 90% of the samples and Class B around 10%. The ...
87
votes
2answers
7k views

How to understand the drawbacks of K-means

K-means is a widely used method in cluster analysis. In my understanding, this method does NOT require ANY assumptions, i.e., give me a data set and a pre-specified number of clusters, k, then I just ...
0
votes
0answers
9 views

Bisecting K-mediods [duplicate]

Is there an algorithm like Bisecting K-mediods and what would its advantages/weaknesses be? It seems to me that it could be used well in combination of Dynamic Time Warping for clustering time ...
1
vote
0answers
45 views

K-means for dataset with 10 million observations and k = 40 using packages “weightedKmeans”

I am trying to employ K means algorithm on a dataset which has 10 million observations. The unique identifier is a 9 digit US Zipcode and data is collected by a bank on its customers (regarding their ...
0
votes
2answers
78 views

Bisecting K-means using Dynamic Time Warping

I'm trying to cluster time series of different length and I came up to an idea to use DTW as a similarity measure, which seems to be adequate, but the thing is, I cannot use it with K-means, since ...
0
votes
1answer
61 views

k-means triangle inequality

This page explains very good how k-means works: http://mnemstudio.org/clustering-k-means-example-1.htm Somewhere I heard that there is some algorithms, which use triangle inequality to speed up the ...
0
votes
1answer
19 views

K-Means Clustering on Distributed System

Can anyone explain how the k-means clustering algorithm converges on distributed systems? It seems that each node in our hadoop cluster would simply find a local optimum. How do we update across ...