Questions tagged [unsupervised-learning]

Finding hidden (statistical) structure in unlabelled data, including clustering and feature extraction for dimensionality reduction.

Filter by
Sorted by
Tagged with
0
votes
0answers
15 views

Can a Gaussian Mixture model be fit with a continuous response variable?

Does the Gaussian Mixture model require binary and multiclass response/target variable (classification), or can the target vector consist of all real numbers (continuous variable, regression)? Why is ...
1
vote
0answers
6 views

Prerequisites/Checks for performing clustering

What are the checks that should be done on our data before performing clustering? Like how to check whether the dataset contains clusters of equal size/density or the clusters present in the dataset ...
1
vote
0answers
28 views

When to use K-Medoids instead of K-means

When it's better to use K-Medoids rather than K-Means? Can anybody give some examples of dataset for the same?
1
vote
1answer
29 views

Limitations of K-Means Clustering [duplicate]

I was going through a document of Western Michigan University to understand the limitations of K-means clustering algorithms. Below is the link: https://cs.wmich.edu/alfuqaha/summer14/cs6530/lectures/...
1
vote
0answers
11 views

How do you train a clustering model?

This should've been a pretty simple question, but I still have a few questions so I decided to bring the discussion here. The thing is, I have a group of products, and the historical dataset looks ...
3
votes
1answer
48 views

Is Anomaly Detection Supervised or Un-supervised?

AFAIK - One way to process data faster and more efficiently is to detect abnormal events, changes, or shifts in datasets. Anomaly detection, also known as outlier detection is the process of ...
0
votes
2answers
40 views

Feature extraction in deep neural networks

From many definitions that I read, I concluded that a DNN (deep neural network) is an ANN (artificial neural network) that have more than one hidden layer. Knowing that CNN (convolutional neural ...
0
votes
0answers
9 views

Gaussian Mixture Model Clustering - cluster means are assigned to a different cluster

I ran a gaussian mixture model with 7 clusters on my data. My data has been PCA transformed with 200 components. Then I extracted the means of each cluster and applied the predict_proba function on ...
2
votes
1answer
21 views

What statistical tests or machine learning techniques would be best for discrete variables with many levels?

I have a dataframe with 20 categorical variables, each with 30+ levels. As a result I don't have a target variable on hand per-say but I would like to use statistical techniques or machine learning to ...
0
votes
0answers
19 views

Why is my training unstable?

I am training a Variational autoencoder with and without data labels. When I use labels (blue line), validation error decreases with epochs but without labels (orange line) the training is unstable. ...
0
votes
1answer
31 views

Does automatic feature selection for clustering helps with finding meaningful clusters?

The objective of clustering is to find interesting groups in data. My question is, whether feature selection can substantially help with this objective. I understand feature selection can remove ...
1
vote
0answers
17 views

How to achieve good clustering results in subsequence time series clustering with DBSCAN?

I want to find patterns in a time series and use clustering for that. Before I cluster, I create subsequences from the time series using a sliding window approach. (STS-clustering) So far I have tried ...
1
vote
1answer
14 views

Differentiability in Generative Adversarial Networks

I've got some questions about the differentiability condition of GAN's, i.e. both G and D need to be differentiable wrt. their inputs and the parameters describing them. It's of more mathematical ...
0
votes
0answers
25 views

stepwise clustering

Assume that we have 3 features in our dataset and we aim to cluster them. Assume that first two variables are in the same scale and have a "similar nature" and the third one has totally different ...
0
votes
0answers
26 views

How to deal with extremely low variance features

I am currently in the process of doing feature extraction for cluster analysis. Some of my features have an extremely low variance, resulting in a very "narrow" distribution, meaning that most ...
0
votes
0answers
6 views

Unsupervised learning over background knowledge

I have a question related to logic programming. Say you have a database of facts like $\{Tall(Jordan), Smokes(Jordan), Tall(Pat), ¬Smokes(Pat)\}$ I want a learning algorithm that will derive from ...
0
votes
0answers
6 views

What is application of topic model?

For example, I have a collection of documents,then I use document-word matrix transformed from these documents to fit model lda(document_word_matrix) by ...
1
vote
1answer
22 views

Is John Skilling's Nested Sampling Algorithm a Supervised or Unsupervised Learning Technique?

Is John Skilling's Nested Sampling a Supervised or Unsupervised Learning Technique? See https://en.wikipedia.org/wiki/Nested_sampling_algorithm.
1
vote
0answers
21 views

Statistical methods to profile clusters

I am working on a clustering exercise and now I need to generate business insights by profiling the clusters(provide a brief description of each cluster. Eg : Cluster 1 mostly consists of high ...
0
votes
0answers
8 views

How to mathematically prove that cost function decrease if the “centroid” is updated in K means?

How to mathematically prove that cost function decrease if the “centroid” is updated in K means?The cost function is : This cost function ,which is the sum of square of the distances of each point to ...
0
votes
1answer
21 views

What happened when k=1 in k means? What's the optimized value of distance for k=1?

What is the optimized value of distance V(x,c) when k=1 (number of cluster) in k means? What is the centroid such that it is optimal? which is the sum of square of the distances of each point to the ...
0
votes
1answer
33 views

What is the mathematical definition of the 'Elbow Method'?

In K-means algorithm, it is recommender to pick the optimal K, according to the Elbow Method. However all the tutorials explain the elbow method in these 4 steps: Run K-means for a range of K's ...
0
votes
0answers
10 views

How do I approach this problem?

Let's say I have a dataset with multiple types of multiple ingredients (salt1,salt2, etc). Each n-th variation of each ingredient vs flavor may be represented by an n×k matrix that where an ingredient ...
2
votes
0answers
24 views

Theoretical justification behind assuming that the data is locally uniformly distributed, as seem to be used by manifold learning community

In at least three or more papers I've been studying that introduced novel algorithms for the estimation of intrinsic dimensionality (ID) based on nearest neighborhood (NN) techniques, I observed that ...
2
votes
1answer
44 views

Does it make sense to do PCA when you got two variables with the same scale?

I want to make to measure perceived standard of living in different countries in Europe. I got two variables from the European Social Survey that almost covers this: slvpen: Standard of living of ...
0
votes
0answers
11 views

Cherry Pick Topics from trained LDA Model

I've got a quick question: Assuming I trained a LDA topic model with K = 150. The majority of the topics looks great but some of them are nonsense or more or less duplicates of other found topics. ...
1
vote
0answers
26 views

Initialisation strategies for learning Hidden Markov Models

I used hmmlearn library to initialize an HMM (Hidden Markov Model). sampled observations from the HMM, and used the sampled data to re-estimate the parameters of ...
0
votes
0answers
16 views

Checking accuracy of classical Multidimensional Scaling - how to define metrics to measure accuracy?

Say I'm given a set of distances between the samples in a data set and a given dimension $p.$ I'm asked to embed the dataset in $\mathbb{R}^p$ using classical multidimensional scaling (MDS). To be ...
1
vote
1answer
20 views

Why do we need Auto-encoder? How is it useful in real life implementation? [duplicate]

From what I've learned about autoencoder is it takes an input and predicts an output almost similar to the input. So, if it outputs the same thing with the same dimensions, what is the benefit of ...
1
vote
1answer
19 views

Anomaly detection on high dimensional Data using k means/SVM/LOF?

I am working on one Anomaly detection problem (unsupervise problem) Data set have 1) 15 columns and around 8k rows , including normal and abnormal(outlier ) rows, without label , all are numeric ...
0
votes
0answers
10 views

Equivalent of NMI and B3 for multilabel extrinsic clustering evaluation metrics

Normalized Mutual Information (NMI) and B3 are used for extrinsic clustering evaluation metrics when each instance (sample) has only one label. What are equivalent metrics when each instance (sample) ...
0
votes
0answers
19 views

How do I evaluate a K-Means unsupervised anomaly detection approach?

how do I evaluate K-means clustering anomaly detection method as there is no labelled data of anomaly class. To find the cluster (K), I have used the silhouette score from Scikit learn library. Scikit ...
1
vote
1answer
29 views

Unsupervised soft clustering methods

I have a D-dimensional dataset composed of exactly two clusters (this is known) for which I have no labels; the clusters can potentially be wildly imbalanced. I'm after a soft (or fuzzy) clustering ...
0
votes
0answers
20 views

How to deal with missing values in Autoencoder model

I have a multivariate time series data set which has a lot of missing values. I'm trying to make an anomaly detection model using autoencoder. I've read this : "https://stackoverflow.com/questions/...
0
votes
0answers
17 views

Kmeans in high dimentions with features with different scales: how to normalize

I am developing an unsupervised clustering approach on a dataset with high variability. The dataset has 3 main characteristics, that makes it a little bit more complicated then others. The #of ...
1
vote
0answers
27 views

One shot inference with Variational Autoencoders using proposal mean

Let's say you have an already trained Variational Autoencoder where the parameters are $\phi, \theta$ for the recognition and generative models respectively. Let's also assume you have the following ...
1
vote
1answer
33 views

Get the eigenvalues when you know the explained variances of a PCA plot

I'm performing a PCA using the sklearn.decomposition.pca function. It appears to work as it should. Acording to this question, I can get the eigenvalues like this:. ...
1
vote
1answer
23 views

Why atoms in the dictionary of Dictionary Learning method are not required to be orthogonal?

According to Sparse Dictionary Learning (wiki), Sparse dictionary learning is a representation learning method which aims at finding a sparse representation of the input data (also known as sparse ...
0
votes
0answers
15 views

What should I be careful when using the word “supervised” in paper writing?

I am a biologist using machine learning tool for my research. I modified matrix decomposition ($V \approx WH$) to fit my data and wanted to describe about that in my paper. If I fixed one matrix ...
0
votes
0answers
18 views

Intertia significantly decreasing after running multiple k-mean clustering models

I have been spending some time running k-means clustering models using scikit-learn on a variety of feature combinations and have been using the inertia value to compare models to one another. I've ...
2
votes
2answers
45 views

Missing data in k-means cluster model

I'm working on clustering email addresses using K-means based on their value to and engagement with the company (metrics such as % of emails opened, # of web browsing sessions, etc). I would like to ...
0
votes
0answers
20 views

Preprocessing on unsupervised learning

I am working on a high dimensional problem that evaluates code readability according to specific metrics. The problem is that there is no 'ground truth' so I need to implement clustering (instead of ...
2
votes
0answers
17 views

Productionize (applied) PCA to detect outlier etc. long term with new data

I was wondering how one could use PCA in e.g. a dashboard for non Subject Matter Expert. For example, you are quite certain that 2 PCs are sufficient based on the current data. It also makes sense ...
0
votes
0answers
9 views

Is kinship detection limited to classification / unsupervised learning?

There are numerous studies and ML approaches published around kinship detection. Generally, the system is presented a pair of inputs (e.g, 2 photos) and a score is produced, enumerating the degree of ...
0
votes
0answers
12 views

What approach or unsupervised methods can be used to pick out patterns in noise?

This is a hypothetical situation. Let's say you have access to a lot of human behaviors and characteristics (features). Let's say you have a sample of 10000 humans. You know that within this sample, ...
0
votes
1answer
31 views

Kernel selection for one-class SVM learning

Has anyone seen compelling research on kernel selection for one-class SVM learning? I've not tracked this work in some time and am wondering if there's new work I've missed, particularly from the ...
0
votes
0answers
14 views

Are there any methods to detect whole multivariate time-series as anomalous from a set of multivariate time-series?

Consider a scenario with Dataset D as {T1, T2, ..., Tn} and Ti is a multivariate time-series of length mi as {X1, X2, ..., Xmi}. Here each record of the time-series Xi is a vector of attribute values {...
0
votes
0answers
21 views

To combine dataset or not to combine dataset in unsupervised clustering?

I have datasets of N different machines of the same type, for example: N = 1000, type = BMW i3. On these datasets I run a unsupervised clustering algorithm. My question is now: Does it make more ...
0
votes
0answers
31 views

How does one calculate maximum dissimilarity?

I created this test Dataset data <- data.frame(Gene1 = c(1,2,1,8,9,7), Gene2 = c(5,6,6,3,4,4), Class = c("Male", "Male", "Male", "Female", "Female", "Female")) ...
3
votes
1answer
27 views

Dimensions Of The Covariance Matrix

I know that PCA can be obtained by eigendecomposition of the covariance matrix, and the covariance matrix $S$ is obtained by the equation: $S = X^TX $, where $X$ is the centered data matrix. But I ...

1
2 3 4 5
12