Hot answers tagged

44 votes
Accepted

Latent Dirichlet Allocation vs Hierarchical Dirichlet Process

HDP is an extension of LDA, designed to address the case where the number of mixture components (the number of "topics" in document-modeling terms) is not known a priori. So that's the reason why ...
user avatar
  • 3,020
28 votes

Latent Dirichlet Allocation vs Hierarchical Dirichlet Process

Anecdotally, I've never been impressed with the output from hierarchical LDA. It just doesn't seem to find an optimal level of granularity for choosing the number of topics. I've gotten much better ...
user avatar
23 votes

Latent Dirichlet Allocation vs Hierarchical Dirichlet Process

I wanted to point out, since this is one of the top Google hits for this topic, that Latent Dirichlet Allocation (LDA), Hierarchical Dirichlet Processes (HDP), and hierarchical Latent Dirichlet ...
user avatar
  • 231
20 votes
Accepted

What does the alpha and beta hyperparameters contribute to in Latent Dirichlet allocation?

The Dirichlet distribution is a multivariate distribution. We can denote the parameters of the Dirichlet as a vector of size K of the form ~$\frac{1}{B(a)} \cdot \prod\limits_{i} x_i^{a_{i-1}}$, where ...
user avatar
  • 4,408
13 votes

What does the alpha and beta hyperparameters contribute to in Latent Dirichlet allocation?

Assuming symmetric Dirichlet distributions (for simplicity), a low alpha value places more weight on having each document composed of only a few dominant topics (whereas a high value will return many ...
user avatar
  • 181
10 votes

Tutorials on topic models and LDA

If you're working in R, Carson Sievert's tutorial on using LDA to model topics in movie reviews is an excellent starting point: http://cpsievert.github.io/LDAvis/reviews/reviews.html This tutorial ...
user avatar
8 votes

Tutorials on topic models and LDA

I highly recommend this tutorial: Getting Started with Topic Modeling and MALLET Here are some additional links to help you get started... Good introductory materials (including links to research ...
user avatar
5 votes

Clustering of documents using the topics derived from Latent Dirichlet Allocation

Assuming that LDA produced a list of topics and put a score against each topic for each document, you could represent the document and it's scores as a vector: ...
user avatar
  • 181
5 votes

Why we should not feed LDA with tfidf

Since the StackOverflow link in the question comments seems broken, here is another reply that addresses the same question: https://stackoverflow.com/a/44789327/6470915 Direct quote: In fact, Blei ...
user avatar
  • 151
4 votes

Why do my Latent Dirichlet Allocation Topics mix words that never co-occurred?

It could be partially due to the number of topics you selected but the fact that two words rank high for a given topic doesn't necessarily mean that the two words will frequently occur in the same ...
user avatar
  • 836
4 votes

Need help with LDA for selecting features

Try before doing LDA look at the data - like doing TF, IDF and TFIDF analysis to identify such words which happen in all subject. If You have some taxonomy of Your product definition - consider using ...
user avatar
4 votes

News topic detection and categorization

I'm not an expert in this field but you should take a look at the work of Bhargav Srinivasa Desikan, a gensim contributor, who works a lot with topic modelling. He has a couple of notebooks on his ...
user avatar
4 votes
Accepted

How to split natural language script into segments?

I haven't seen anything like this before but it seems quite feasible. You need an ontology to separate the main concept into its subconcepts, then you need a classifier to distinguish between your ...
user avatar
  • 10.4k
4 votes
Accepted

What's beyond topic modeling?

You could use doc2vec to create vector representations of each document. Once you have all the vector representations you can use standard unsupervised clustering techniques like k-means, hierarchical ...
user avatar
4 votes
Accepted

Calculating optimal number of topics for topic modeling (LDA)

LDA being a probabilistic model, the results depend on the type of data and problem statement. There is nothing like a valid range for coherence score but having more than 0.4 makes sense. By fixing ...
user avatar
  • 1,246
3 votes
Accepted

replicability / reproducibility in topic modeling (LDA)

LDA is Bayesian model. This means the desired result is a posterior probability distribution over the random vectors of interest (probability of topics etc. having seen some data). Inference for ...
user avatar
3 votes

Tutorials on topic models and LDA

If you are looking for something simple to start with and easy to implement, I would recommend this. Beginners Guide to Topic Modeling
user avatar
  • 271
3 votes

Need help with LDA for selecting features

I think, even before doing LDA, you should remove words which appear in more than "x" percent of your documents. Try different "x" starting from 80% and then going down. The logic is that if the word ...
user avatar
  • 1,446
3 votes

Goodness of fit metric to compare topic models NMF vs LDA

NMF and LDA models produce topic-word and document-topic distributions, so you can compare these models on evaluation tasks of topic coherence (i.e. evaluating topic-word distributions), document ...
user avatar
  • 211
3 votes
Accepted

Equally sized topics in Latent Dirichlet allocation

It is normal, the best explantation I found for it, is from physics. Since Gibbs Sampling was known in physics long before LDA and LDA can simply be seen as a kind of matrix factorization. There is a ...
user avatar
  • 445
3 votes
Accepted

k-means and LDA for text classification: how to test accuracy?

First of all, you use two terms Clustering and Classification interchangably and I would like to draw your attention to this. Your problem is purely Clustering. Secondly, you asked for testing ...
user avatar
3 votes
Accepted

TF-IDF for Topic Modeling

Formally the problem of topic modelling is a clustering problem: given a collection of text documents, group together the documents which are topically similar. So technically it can indeed be done ...
user avatar
  • 20.8k
2 votes

Latent Dirichlet Allocation vs Hierarchical Dirichlet Process

Yee Whye Teh et al's 2005 paper Hierarchical Dirichlet Processes describes a nonparametric prior for grouped clustering problems. For example, the HDP helps in generalizing the Latent Dirichlet ...
user avatar
  • 2,301
2 votes

Tutorials on topic models and LDA

I suggest trying Machine Learning Plu's Gensim tutorial. It will give you a holistic overview, on NLP and LDA, including: how to pre-process your data, do feature engineering and apply LDA.
user avatar
  • 21
2 votes

Proceeding with various methods for news recommendation

Recommender Systems are a huge topic of its own right and goes without saying, with a lot of research going on. This book does a deep-dive into recommender systems and may not be something you want, ...
user avatar
  • 141
2 votes

How to compare LDA and TF-IDF?

If you have the ground truth value of the documents (their topics) all you gotta do is pick a metric and compare results. For classification problems, as yours, a common metric would be f1_score; ...
user avatar
  • 166
2 votes
Accepted

Can I use euclidean distance for Latent Dirichlet Allocation document similarity?

Euclidean distance -by which in this application, I assume you mean the euclidean distance in an $n$-dimensional space defined by the distribution of document contents among $n$ topics considered, is ...
user avatar
2 votes

Topic modeling for short length sentences

A possible approach might be to use the most predictable and predictive word(s) in each cluster as its name(s). The following is inspired by Fisher's category utility used by the COBWEB algorithm. An ...
user avatar
  • 276

Only top scored, non community-wiki answers of a minimum length are eligible