Stack Exchange Network

Stack Exchange network consists of 175 Q&A communities including Stack Overflow, the largest, most trusted online community for developers to learn, share their knowledge, and build their careers.

Visit Stack Exchange

Questions tagged [unsupervised-learning]

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

1
vote
1answer
572 views

Unsupervised learning methods on unlabeled data?

I'm facing with a challenge of unsupervised classification of unlabeled data. The case is, I have circa 1.2 million vehicle warranty claims, and must develop a classification model to tell whether ...
6
votes
4answers
666 views

t-SNE dimensions as additional predictor variables

This question could also (maybe) relate to PCA. I built a supervised RandomForest on a dataset that I'm currently working on - the actual V Prediction $R^2$ was holding around 80% across many CV ...
2
votes
0answers
71 views

Visualizing neural network inferences

I know this is an ongoing and hard question to answer, but if anyone has experience in this then please share your knowledge. Suppose I have made a neural network with the task of predicting an ...
0
votes
1answer
352 views

One-class SVM: “training set vs. origin” logic

first of all, I apologize about asking this question again since a similar one was posted recently; I had to repost it since I still don't understand the answer and I had no other way to interact with ...
0
votes
1answer
89 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 ...
2
votes
0answers
124 views

Random forest - extracting profiles

I'm using the good old decision tree (CRT or CHAID algorithm, depending on the situation) in order to predict voting behavior and extract some profile (e.g: Women who live in the suburbs, who are not ...
0
votes
0answers
63 views

What is a medium-sized dataset for scikit learn clustering algorithms?

On scikit learn's clusting algorithm page they give a comparison chart for different clustering algorithms describing their scalability and usecases. What is considered a medium sized n_samples? I ...
0
votes
3answers
401 views

Prediction after PCA and K-Means

I have a data set with a large amount of features. I'm applying PCA on it in order to run it through K-means, to discover clusters in my data set. I'd like to know what is the best practice to make ...
3
votes
1answer
2k views

Unsupervised training of CNN

I have some unlabeled 1D (i.e. time-domain) signals (real neuron measurements) that I would like to classify in 3 classes. I would like to use a ConvNet to do this. However, as far as I know, ConvNets ...
1
vote
1answer
233 views

Supervised, semi-supervised, or unsupervised? Confused

BACKGROUND: Consider the problem in cybersecruity that consists of classifying domain names as either malicious or legitimate based on various features such as the URL string, the name of the ...
2
votes
1answer
124 views

what are the top level subsets/domains of ML?

I'm not really happy with the mind maps I've been able to find on Google, most of them are algorithm based. I want to make a good one that is problem/solution domain based. Do I have this right for ...
2
votes
1answer
312 views

Clustering before regression

Could there be any benefit to running a clustering algorithm on a data set before performing regression? I'm thinking that it might be useful to run a regression algorithm on each cluster thereby only ...
2
votes
0answers
260 views

State-of-the-art: unsupervised learning for patterns in text

My gut tells me tried-and-true approaches like k-means and Latent Dirichlet Allocation may no longer be state-of-the-art approaches for unsupervised learning with text data, what with models like ...
1
vote
0answers
136 views

Prediction using Hidden Markov Models

I am trying to model financial time series data using Hidden Markov Models. This question is related to time series analysis in general. Can I create the model on previous few days data and use that ...
1
vote
0answers
57 views

How does using a correlated topic model effect the distribution of topics?

Correlated Topic Models are a great advance on the original topic model - see Blei and Lafferty 2007 for more info. My question is this - how does a Correlated Topic Model impact the overall ...
0
votes
0answers
690 views

Learning normal distribution with VAE

I am trying to use a Variational Autoencoder to learn a multivariate normal distribution. I know that from practical point of view this is pointless, as we can sample from a normal distribution itself,...
3
votes
1answer
390 views

Which methods can help us to understand clustering model is good or bad?

In some clustering algorithm, ex: K-Means cluster, it is very sensitive with outliers, so we need to remove outliers before aplly ...
2
votes
0answers
306 views

Clustering of high dimensional data

I am having a data set with 54 independent variables .Most of them are having zeros it resembles like sparse matrix .How to cluster this kind of data and is there any data pre processing like Box cox ...
1
vote
0answers
176 views

Similar to Self Organization Maps, are all other clustering algorithms “self organizing”?

From my understanding of the terms: Self means: No supervision is required during training. Organizing means: To create a topographic ordered map by using unsupervised competitive learning and ...
3
votes
1answer
153 views

Does online learning theory have any real world applications?

This is a question regarding the specific application of online learning theory in the sense of http://www.mit.edu/~9.520/spring08/Classes/online_learning_2008.pdf I went through ICML papers for 2017 ...
2
votes
0answers
286 views

How to do model selection in the unsupervised learning setting?

For supervised learning, we know the correct answers for samples, model selection is more easier, we can use k-ford cross validation (this site!) and etc. But for unsupervised learning, e.g. ...
3
votes
0answers
194 views

Topological data analysis and evaluating dimensionality reduction

I did an exploration some time ago on using TDA tools to see how topological features change after application of some nonlinear dimensionality reduction methods. For example I found out that, for ...
0
votes
0answers
50 views

How can I use the results of clustering algorithms for classification

I'm doing a mobile customer segmentation and I was using K-means to cluster my data according to the various data points (location, time of use, duration used for etc). After reading a lot of posts in ...
0
votes
1answer
385 views

How to validate k-means result [duplicate]

I'm doing anomaly detection on unsupervised data using k-means I got a result but I don't know how to validate my clustering result. by plotting I can see my anomalies but how should I validate that ...
0
votes
1answer
238 views

Threshold for kmeans anomaly detection

I'm learning the kmeans to find out anomaly from the dataset. but I don't know how to set threshold. I tried by the putting mean of the centroid to point distance but it's not working, half my record ...
0
votes
1answer
3k views

how to handle sparse data problem in unsupervised learning .i'm going to use k means on data set

how to handle sparse data problem in unsupervised learning .i'm going to use k-means on the dataset. I have 200 variables, nearly in each column have 70% zeros. how can I handle without discarding any ...
0
votes
1answer
46 views

How to improve classification algorithms when the features are unlabeled?

hey guys I'm dealing with data that has around 15 features that are all positive, real numbers but I have no idea what the features are. The headers of my data are all just "feature1", "feature2", "...
0
votes
1answer
267 views

What is a good reason why reinforcement learning relies on Q value instead of the reward r?

I am new to reinforcement learning and I am reading off of some tutorial materials, but I have noticed a seldom discussed assumption that we should calculate our action based on Q-value instead of the ...
0
votes
0answers
60 views

How can I classify my text by using unsupervised approaches?

I have a corpus of newspaper articles and would like to classify them according to content (words, phrases) that I determine beforehand. The only way to do this that I found so far is by looking at ...
0
votes
1answer
55 views

Learning vs. training

In my head, these two words (learning and training) seem to somehow have a fuzzy boundary between them. For example, the word learning for me conveys the idea of training; if I want to learn ...
1
vote
0answers
33 views

How can we learn the values of the parameters for Levenshtein distance?

I am trying to filter out similar-looking names from a database. Once I have figured out the names, I will merge them into a single entity. To achieve this, I am planning on using edit-distance ...
1
vote
1answer
2k views

how to handle missing data in clustering problem

The features in data sometimes contains missing values, which mean the value is unknown. If I replace unknown value with a special normal value like "0", then the clustering algorithms will trade them ...
0
votes
1answer
35 views

What is the best way to analyze and predict based on a dataset that has both text and numbers?

I have a Twitter dataset that has both the tweets themselves, but also metadata about the user as number of tweets, date of creation, and other numerical values. How can I make predictions based on ...
2
votes
0answers
54 views

How can knowing test data improve the performance of face recognition?

I don't think knowing face recognition is necessary to answer this question, any suggestions are welcome. For face verification problems, usually images are first converted into fixed length vectors ...
1
vote
1answer
30 views

Improving clusters

Say we have 10 classes. We have obtained two kinds of clusters in an unsupervised manner. One type is heavily clustered, i.e, say it shows only 4 clusters and one heavily cut, i.e, it shows say, 20 ...
5
votes
5answers
470 views

Choosing a model for my unsupervised machine learning problem

I need to choose a model for unsupervised machine learning problem. There are 4 clusters in 3D space. These are my requirements: I will run the same model multiple times with different training data (...
0
votes
1answer
133 views

Resulting clusters are very uneven

I am clustering a set of 50k products. I would expect the resulting clusters to be things like "organic chicken", "orange juice", etc. I am using the bag of words model, and there are about 8k ...
0
votes
1answer
292 views

Cluster data based on similarity, but split by one feature (Python, R)

Is there a way to do unsupervised clustering based on similarity (like all other methods), but create clusters by splitting on just one (or specific) features? For example, I have customer data with ...
2
votes
1answer
101 views

Any way to recognize pattern(such as Char and Number) from image without labeled data?

I am trying to build a captcha recognizer. I found CNN play very well if there are enough labeled data. For example, I use this https://github.com/lepture/captcha to generate 4 size char+number ...
2
votes
1answer
2k views

Which unsupervised classification method can be used for categorical data?

I have a list of categorical data and I want to apply an unsupervised classification method to cluster this data. Which method could be used? Example: gene1 gene2 gene3 gene4 ...
1
vote
1answer
214 views

What is the difference between SVI and online HDP algorithms?

I'm going over the paper by Hoffman et. al. titled "Stochastic Variational Inference" (SVI) where the authors present an algorithm for Hierarchical Dirichlet Process (HDP) topic models. I'm comparing ...
1
vote
1answer
421 views

clustering for categorical data with one column for observations

I'm trying to cluster a dataset using 4 variables, all of which are categorical variables. I'd also like to include another numerical variable that's actually the number of observations of another ...
1
vote
1answer
121 views

Objective performance measures for PCA?

How can I measure how good my PCA was ? Of course, it depends A LOT on context, or even maybe subjectivity sometimes. But what objective measures can I use to measure how good my PCA was? It doesn't ...
4
votes
1answer
2k 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 ...
0
votes
1answer
61 views

Branches or taxonomy of dimensionality reduction?

I need a high-level taxonomy of dimensionality reduction approaches. I need your help to unify what I have found: 1) according to wikipedia, dimensionality reduction can be divided into: feature ...
0
votes
1answer
38 views

Can we say that clustering is a function?

Can we say that clustering is a function? Of course, the term "clustering" is broad and can mean different things in different contexts: clustering as an area in machine learning, clustering ...
1
vote
1answer
221 views

What is the typical taxonomy for clustering methods?

What is the typical taxonomy for clustering methods? For example, for regression we can talk about: simple regression, multiple univariate regression, and multivariate regression. And then, we can ...
5
votes
1answer
2k views

Generative models for time series simulation

I have a basic understanding of generative models, like Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs). It seems that they are mostly used to in the field of image ...
2
votes
0answers
277 views

Metric for unsupervised recommender-system competition?

I have a data source containing millions of documents from a wide variety of business domains. We've aggregated the data such that we can easily find information using natural-language search queries. ...
0
votes
0answers
53 views

“Uniform” Clustering

I have the data $\{(y_i,\omega_i), i=1,\dots,n\}$, where $y_i \in \mathbb{R}_+$ the response and $\omega_i>0$ a weight. Fixed $K>0$ I want to determinate $y_1^*,\dots,y_{K-1}^*$ such that $$ \...