WenRichard/KBQA-BERT
A knowledge-base question answering system that fine-tunes BERT for named entity recognition and sentence similarity matching.
Not currently ranked — collecting fresh signals.
star history
This project builds a knowledge-graph-based QA system using BERT as its core NLP engine. It performs two main tasks: BERT-based named entity recognition to identify entities in user queries, and BERT-based sentence similarity computation to match questions against stored patterns. The system operates in both online and offline prediction modes, with the final KBQA module integrating both components to retrieve answers from a knowledge base.
Frequently asked
- What is WenRichard/KBQA-BERT?
- A knowledge-base question answering system that fine-tunes BERT for named entity recognition and sentence similarity matching.
- Is KBQA-BERT open source?
- Yes — WenRichard/KBQA-BERT is open source, released under the MIT license.
- What language is KBQA-BERT written in?
- WenRichard/KBQA-BERT is primarily written in Python.
- How popular is KBQA-BERT?
- WenRichard/KBQA-BERT has 1.5k stars on GitHub.
- Where can I find KBQA-BERT?
- WenRichard/KBQA-BERT is on GitHub at https://github.com/WenRichard/KBQA-BERT.