Questions tagged [self-organizing-maps]

SOM is a kind of neural network used for clustering unlabeled data.

Filter by
Sorted by
Tagged with
58
votes
2answers
97k views

How can an artificial neural network ANN, be used for unsupervised clustering?

I understand how an artificial neural network (ANN), can be trained in a supervised manner using backpropogation to improve the fitting by decreasing the error in ...
34
votes
3answers
11k views

(Why) Has Kohonen-style SOM fallen out of favor?

As far as I can tell, Kohonen-style SOMs had a peak back around 2005 and haven't seen as much favor recently. I haven't found any paper that says that SOMs have been subsumed by another method, or ...
23
votes
4answers
7k views

Why are mixed data a problem for euclidean-based clustering algorithms?

Most classical clustering and dimensionality reduction algorithms (hierarchical clustering, principal component analysis, k-means, self-organizing maps...) are designed specifically for numeric data, ...
15
votes
4answers
26k views

Text Mining: how to cluster texts (e.g. news articles) with artificial intelligence?

I have built some neural networks (MLP (fully-connected), Elman (recurrent)) for different tasks, like playing Pong, classifying handwritten digits and stuff... Additionally I tried to build some ...
11
votes
5answers
2k views

SOM clustering for nominal/circular variables

Just wondering if anyone is familiar with clustering nominal inputs. I've been looking at SOM as a solution but apparently it only works with numerical features. Are there any extensions for ...
10
votes
1answer
196 views

How to automatically cluster a U-Matrix?

After training a self-organising map, one can calculate the U-Matrix. There are some tools to manually visualize it and identify clusters, but I'm wondering if there is any algorithm to do this ...
8
votes
2answers
2k views

Using self organizing maps for dimensionality reduction

Over the past few days, I have been conducting some research on self organizing maps for a project at school. I have come to understand that self organizing maps can be used to reduce the ...
8
votes
1answer
326 views

Self-organizing maps: fuzzy input?

I would like to know if there are SOM implementations (preferably R) available that accept fuzzy input. That is, I have data in which some nominal features are spread out between a number of ...
7
votes
2answers
4k views

How do SOMs reduce dimensionality of data?

This is a problem with which I have been grappling with for days. From my research on self-organizing maps, I know that a common feature of self organizing maps is to reduce the dimensionality of data....
6
votes
3answers
10k views

Self organizing maps vs k-means, when the SOM has a lot of nodes

On Wikipedia it says: It has been shown that while self-organizing maps with a small number of nodes behave in a way that is similar to k-means, larger self-organizing maps rearrange data in a way ...
6
votes
4answers
3k views

Dimensionality reduction using self-organizing map

Self-organizing maps are claimed to be an approach for dimensionality reduction. However, I am kind of confused about this claim. Consider the following example, I have a data set with 200 data ...
5
votes
2answers
2k views

How do I compare multiple runs of K-means?

I have results of best centroids for multiple (10) runs of k-means. How do I compare these weights to check if they are close to each other or different? My goal is to check weather I get to the ...
5
votes
1answer
3k views

SOM grid size suggested by Vesanto

I am a bit confused on the size of the SOM grid size suggested by Vesanto. Here in this link, it says 5*sqrt(N) where N is mentioned as the dataset size. What is ...
5
votes
1answer
2k views

What is the computational complexity of the SOM algorithm?

Assuming $m$ observations, $n$ features and $k$ nodes in the self organizing map, what is the complexity of the classic SOM algorithm? What would be the complexity of an ensemble of SOMs, where each ...
5
votes
1answer
1k views

Difference between SOM and Hopfield

I am learning about SOM and also about Hopfield networks. I yet don't understand why I would use SOM over Hopfield and vice ...
5
votes
1answer
188 views

How much do self organising maps suffer from local minima problems?

How much do self organising maps suffer from local minima problems? I assume that the original orientation of the grid can have major orientation on the resulting shape of the grid, but does this have ...
5
votes
0answers
718 views

Differences between t-SNE and SOM

I have some high dimensional data and I want to reduce it to 2 dimensions for visualization. The goal is to color the points in this 2D space to see whether there is any clustering due to different ...
4
votes
1answer
561 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 ...
4
votes
1answer
718 views

Growing hierarchical self-organizing maps

For learning GHSOM I figured out I should study SOM as a first step. Now, I know the basics of SOM, about weighted vectors and euclidean distances: when the human brain cannot easily process more ...
4
votes
0answers
1k views

What is the main differences between choosing hexagonal grid and rectangular grid for SOM?

While I'd expect people to answer this question by saying 'depends on the distribution of data', but what are the thumb rules for deciding which grid to use (either hexagonal or rectangular) for ...
4
votes
0answers
389 views

Kohonen SOM for high (50-100) dimensions

Does a Kohonen-style SOM, using Euclidean distance, work as well as, better than, or worse than alternatives (K-means, etc) in high (50-100 or more) dimensional space? EDIT: I'm thinking particularly ...
3
votes
2answers
1k views

How does one visualize the self-organizing map of $n$-dimensional data

I have a data set consisting from $7$-dimensional data points. I want to produce a self-organizing map for this data to see how my data is clustered. I have been reading some tutorials from the ...
3
votes
1answer
446 views

Self Organizing Maps: How is the location computed and updated?

I have read other similar questions on here, but I am still unsure how SOM deals with the positions/locations of the neurons. Say that the input space is N-dimensional. I initalise some weights, and ...
3
votes
1answer
2k views

How to set the radius value in self organizing map?

I'm training the self organizig map, I need to set the value for the radius of it. is there any method to find the optimum radius size ?
3
votes
1answer
787 views

Evaluating the clustering of a Kohonen UMatrix

Given a converged Kohonen feature map, how would one evaluate the clustering in terms of intra- and inter-cluster distances? Assuming that both the trained codebook vectors and Unified Distance ...
3
votes
1answer
642 views

Kohonen SOM in R: How to give weights for certain variables in the BMU finding process?

I'm using the Kohonen package (see also self-organising-maps-for-customer-segmentation-using-r) for Self Organizing Maps (SOM), and I would like to know how to give weights for certain variables in ...
3
votes
2answers
2k views

Kohonen self organizing maps: determining the number of neurons and grid size

I have a large dataset I am trying to do cluster analysis using SOM. The dataset is huge (~ billions of records) and I am not sure what the number of neurons should be or the SOM grid size to start ...
2
votes
1answer
705 views

Are Self organizing maps networks a variant of Multi Layer Perceptron?

I'm learning Self Organizing Map (SOM) networks and I think I can say that SOM networks are a Multi Layer Perceptron (MLP) network. Is it correct to say that SOM is a variant of MLP?
2
votes
1answer
2k views

Using the self-organizing map for sequences of categorical data

I have a number of vectors of categorical data (ex. {'re','ty','cf', ...} ) and I want to perform an unsupervised learning on them. I came across self-organizing map...
2
votes
1answer
1k views

Self Organizing Map and input normalizing

I've been playing around with self organizing maps (SOM) recently. I tried to implement a simple example. You can see the training implementation function gist here and full contained SOM example ...
2
votes
1answer
414 views

Growing Self-Organizing Map for mixed-type data

I am trying to write code to build a growing SOM for mixed-type data. I came across a paper Growing Self-Organizing Map with cross insert for mixed-type data (http://www.sciencedirect.com/science/...
2
votes
1answer
330 views

Dimensionality reduction with self organizing map

Suppose we have 20 training data points in 50 dimensions. Let's say I have specified a 3 by 3 SOM (lattice with 9 points), I embed my manifold (3 by 3 lattice) to 50-D space and after training each ...
2
votes
1answer
851 views

Why SOM is better than clustering technique(e.g. hierarchical)?

I am using SOM for dimension reduction and visualization purpose (to put the same observations together). I am using kohonen r-package for the same. https://cran....
2
votes
1answer
260 views

Why is SOM not being used? [duplicate]

What SOM does sounds good and very useful on a paper (putting similar individuals close together, nonlinear version of PCA) for visualization and for dimensional reduction. Also there are whole ...
2
votes
1answer
1k views

Self-Organising Maps and missing data (NAs) in R

The SOM algorithm should be able to deal with some datapoints containing NAs: to find their Best Matching Units, it would be possible to compute Euclidean distances with the neurons ignoring the ...
2
votes
1answer
793 views

self-organizing maps and order of samples

I've just started learning about self-organizing maps without much of a background in neural networks, so forgive me if this question seems trivial, but it seems that a SOM depends on which order the ...
2
votes
1answer
850 views

Can we use self organizing maps to build a reconstruction model such as autoencoder or generative adversarial network?

MLP and its variants can be used to build a reconstruction model such as conventional autoencoders including many variants and GAN, for example. I wonder if SOM can be a candidate to replace the ...
2
votes
1answer
323 views

Empty nodes when creating SOM

I am trying to create a SOM map based on records with different discrete classifications (tags) like the example below ...
2
votes
1answer
100 views

How does dimensionality of network topology affect self organising maps?

I'm pretty new to self organising maps (SOMs). From reading around a bit, it seems like most implemetations use a 2D grid for the SOM network structure (or less often a 1D grid). This is nice, cause ...
2
votes
2answers
283 views

Non-linear projection in self organizing maps

I have difficulty understanding how self organizing maps (SOM) are doing dimensionality reduction. Can anybody provide a useful explanation to me? Suppose we have 20 training data points in 50 ...
2
votes
0answers
426 views

Choosing a Sample Size for a SOM Cluster

Is there an accepted formula to determine a good lower bound for the number of samples needed for the initial input on an SOM clustering algorithm? For example - If I know that there are X number of ...
2
votes
2answers
2k views

How to interpret “weight-position” plot when using self-organizing map for clustering?

I used MATLAB neural network toolbox to train a self-organizing map for a given data set. The obtained "weight-position" plot is given as follows. I do not think this plot looks good in comparison to ...
1
vote
2answers
606 views

Is it feasible to predict a numerical variable with only 100 measures in 70 000 observations?

A quick explainer: I have a data set of 5 numeric variables, and a numeric target variable. The total observations / rows are about 70 000, but I only have about 90 measurements of the target ...
1
vote
2answers
1k views

SOM based on a not euclidean distance

Suppose one has trained a SOM on a certain number of data. Without explaining all the procedure, one can say that the SOM algorithm produces a certain number of prototypes and the new elements coming ...
1
vote
2answers
335 views

SOM automated/objective clustering

So as I understand it SOM is primarily a visualization tool and clustering is a logical next step after you construct a SOM from data. Typically, the clustering is subjective in that after looking at ...
1
vote
1answer
483 views

SOM (Kohonen) using the term document matrix [closed]

Language: R Package: kohonen Function: som I have a term document matrix (tdm) with 64 terms (row) and 1017 documents (columns). I want to use the self-organized-map to cluster the terms on 25 ...
1
vote
1answer
893 views

Dimension reduction issues in self-organizing maps (SOM) [duplicate]

Self organizing maps are claimed to be able to visualize/cluster high-dimensional data in a smaller dimensional space. I have some difficulties in understanding this statement. Consider a six-...
1
vote
1answer
129 views

Quantify similarity between self-organizing maps (SOMs)

What would be a valid similarity measure to quantify the (dis)similarity between two different datasets processed using the same trained version of a self-organizing map (trained on the combination of ...
1
vote
1answer
250 views

Why SOM shows the best accuracy at 11x11?

I have started working on Human Action recognition using depth images. I found this article. In the experiment section, they have told that they have experimented SOM of different size and for the ...
1
vote
2answers
1k views

Gaussian neighborhood function and non linear learning rate for self-organizing map in R

I've been working on SOMs and how to get the best clustering results. One approach could be to try many runs and choose the clustering with the lowest within sum of squared errors. However, I do not ...