Detecting Redundant Health Survey Questions by Using Language-Agnostic Bidirectional Encoder Representations From Transformers Sentence Embedding: Algorithm Development Study
This study introduces the SBERT-LaBSE algorithm for calculating the semantic similarity across 2 languages, showing that it outperforms BERT-based models, the GPT-4o model, and the bag-of-words approach, highlighting its potential in improving the semantic interoperability of survey-based PGHD across language barriers.
Authors
- Hyeoneui Kim (ORCID: https://orcid.org/0000-0002-5931-7286)
- Ricky K. Taira (ORCID: https://orcid.org/0000-0001-9754-8993)
- Sunghoon Kang (ORCID: https://orcid.org/0009-0003-8133-0895)
- Hye-Won Park
Institutions
- University of California, Los Angeles (US)
- New Generation University College (ET)
Publication Details
- Journal
- JMIR Medical Informatics
- Published
- 2025-04-20
- DOI
- https://doi.org/10.2196/71687
- Primary Topic
- Topic Modeling
- Type
- article
- Field-Weighted Citation Impact
- 0.00