MaartenGr / MaartenGr/BERTopic
UMAP-n_components
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 7.8k
- Forks
- 920
- Avg merge
- 22h 24m
- Merged PRs (30d)
- 5
Description
When I used bertopic to process comment data (more than 90,000 comments), I found that when using the default dimension of UMAP to 5 dimensions, even if I adjusted other parameters, a large amount of useful text was classified as outliers (maybe it was my data problem). When I increased the dimension reduction, the situation would be better. Is it reasonable for me to change the dimension reduction next?Thanks!
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
Start with the BERTopic and UMAP configuration described in the issue, focusing on the default n_components value of 5 and the reported outliers in a dataset of more than 90,000 comments. Reproduce the behavior if the dataset or a comparable example is available, then determine whether higher-dimensional reduction is expected and what guidance or change would resolve the question.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100