Abstract
Automated Frequently Asked Question (FAQ) retrieval provides an effective procedure to provide prompt responses to natural language based queries, providing an efficient platform for large-scale service-providing companies for presenting readily available information pertaining to customers’ questions. We propose DTAFA, a novel multi-lingual FAQ retrieval system that aims at improving the top-1 retrieval accuracy with the least number of parameters. We propose two decoupled deep learning architectures trained for (i) candidate generation via text classification for a user question, and (ii) learning fine-grained semantic similarity between user questions and the FAQ repository for candidate refinement. We validate our system using real-life enterprise data as well as open source dataset. Empirically we show that DTAFA achieves better accuracy compared to existing state-of-the-art while requiring nearly 30× lesser number of training parameters.- Anthology ID:
- 2021.sigdial-1.44
- Volume:
- Proceedings of the 22nd Annual Meeting of the Special Interest Group on Discourse and Dialogue
- Month:
- July
- Year:
- 2021
- Address:
- Singapore and Online
- Venue:
- SIGDIAL
- SIG:
- SIGDIAL
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 423–430
- Language:
- URL:
- https://aclanthology.org/2021.sigdial-1.44
- DOI:
- Cite (ACL):
- Haytham Assem, Sourav Dutta, and Edward Burgin. 2021. DTAFA: Decoupled Training Architecture for Efficient FAQ Retrieval. In Proceedings of the 22nd Annual Meeting of the Special Interest Group on Discourse and Dialogue, pages 423–430, Singapore and Online. Association for Computational Linguistics.
- Cite (Informal):
- DTAFA: Decoupled Training Architecture for Efficient FAQ Retrieval (Assem et al., SIGDIAL 2021)
- PDF:
- https://preview.aclanthology.org/ingestion-script-update/2021.sigdial-1.44.pdf