BERTopic

Research applicability

BERTopic uses transformer embeddings and class-based TF-IDF for topic modelling of document collections.

Getting started

Follow the documented small-corpus quick start; inspect representative documents and outliers before interpreting labels.

Maintenance

Current operational availability and maintenance were not tested; this recommendation rests on the dated source snapshot.

License

Licensing, service terms and rights to input data must be checked before adoption; no independent license audit performed.

Workflow & human oversight

Resource-level classification is intentionally unspecified; instantiated workflows may differ. Humans define objectives, control data access, and check outputs.

Workflow level is not an evidence score; higher is not better evidence.

Approach

The documented pipeline combines embeddings, dimensionality reduction, clustering and topic representations; users can inspect document assignments and customise components.

Limitations

Topic clusters are analytical aids, not validated interpretive themes by themselves. Embedding choice, language coverage, outliers and preprocessing require sensitivity checks; no effectiveness claim is reproduced.

Reproducibility

Reviewed a stored primary-source extract accessed 2026-10-15. No software, model or claimed outcome was independently reproduced.

Adoption requirements

Python environment, a lawful text corpus and expertise to interpret and validate topic groupings.