MaartenGr / MaartenGr/BERTopic
Suggestion: add possible solution for too few topics in FAQ
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 7.8k
- Forks
- 920
- Avg merge
- 22h 24m
- Merged PRs (30d)
- 5
Description
After having the problem of only very few (two) topics from some of my admittedly small data sets, I found that some of them had multiple short, duplicate and uninformative documents.
For example, when fitting to a set of survey responses, documents such as "Don't know", "Nothing to say", "No comment" appeared numerous times in the corpus - and generated one of two topics. Removing these duplicate and uninformative documents led to the generation of 11 topics instead of just two, on a corpus of <500 documents, using the default nr_topics="auto" parameter.
I think it would be helpful to add this (removal of recurrent, uninformative documents) as a possible solution to the FAQ under ["I have only a few topics, how do I increase them?"](https://maartengr.github.io/BERTopic/faq.html#i-have-only-a-few-topics-how-do-i-increase-them)
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Open the FAQ at faq.html and find “I have only a few topics, how do I increase them?”. Read the surrounding guidance, then add recurrent, uninformative documents as a possible solution using the examples from this issue; done when the suggestion appears clearly in that FAQ section.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- documentation
- Issue type
- Documentation
- Difficulty
- 1/5
- Estimated time
- Under an hour
- Activity status
- Stale
- Clarity
- Clearly specified
- Newbie friendliness
- 48/100