# MaartenGr/BERTopic

Leveraging BERT and c-TF-IDF to create easily interpretable topics.

Repository: https://github.com/MaartenGr/BERTopic
Canonical: https://ross.abutalabs.com/products/bertopic
Homepage: https://maartengr.github.io/BERTopic/
Language: Python
License: MIT
License Family: permissive
Topics: bert, transformers, topic-modeling, sentence-embeddings, nlp, machine-learning, topic, ldavis, topic-modelling, topic-models
Last push: 2026-08-22T06:36:06+00:00

## Health v2 (maintenance only)
Score: 77/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 35, longevity 100
- inputs: {"age_days": 2171, "days_push": 11, "days_rel": 273, "gap_med": 129.0, "n_releases_24m": 5}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 7804, forks 917 (observed 2026-08-28T04:10:05.144236+00:00)

## What it is
BERTopic is a Python library for topic modeling that leverages transformer-based sentence embeddings and class-based TF-IDF (c-TF-IDF) to discover easily interpretable topics in document collections. It supports guided, supervised, hierarchical, dynamic, online, multimodal, zero-shot, and LLM-enhanced topic modeling variants.

## Use cases
- discover topics in a corpus of documents
- cluster customer reviews into themes
- track how topics evolve over time
- visualize topics like pyLDAvis
- do zero-shot topic modeling with seed keywords
- summarize topic descriptions with LLMs
- model topics from images
- incrementally model topics in a stream of documents

## When to choose
- you need interpretable topics from text using modern transformer embeddings
- you want flexible topic modeling modes (supervised, dynamic, hierarchical, online)
- you are working in Python and want an actively maintained, well-documented topic modeling library

## When to avoid
- you need a lightweight classic LDA implementation without transformer dependencies
- you work outside Python or need a hosted no-code service
- your corpus is tiny and embedding models would be overkill

## Facets
- artifact type: library
- maturity: stable
- function: nlp, machine-learning, data-science, data-visualization
- domain: machine-learning, data-science
- platform: python, cross-platform
- tags: topic-modeling, bert, transformers, sentence-embeddings, ctfidf, clustering, unsupervised-learning, natural-language-processing

## Member repositories
- MaartenGr/BERTopic (main) score 77

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:05.144236+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:34:37.026577+00:00, confidence not recorded.
  - readme: https://github.com/MaartenGr/BERTopic (fetched 2026-08-28T04:10:05.144236+00:00, sha 580b203c8a04)
  - homepage: https://maartengr.github.io/BERTopic/ (fetched 2026-08-29T08:30:46.751540+00:00, sha ddf33edda9a4)
  - registry_pypi: https://pypi.org/pypi/bertopic/json (fetched 2026-08-29T08:30:46.760856+00:00, sha 015368be8a95)
- Data as of 2026-08-30T08:39:29.467469+00:00.