Research
Ontology Alignment with LogMapBio+
Research project · 2024 – 2025
A two-phase alignment pipeline. LogMapBio produces the initial mappings, and SBERT transformer models validate them semantically, catching mappings that string matching misses.
- maximum recall
- 92.48%
- correct mappings recovered that lexical methods missed
- 826+
- reduction in processing time from parallel execution
- 70%
- total runtime across 24,000+ entities
- 38 min
Problem
Large biomedical ontologies (DOID, NCIT, FMA) use different vocabularies. Lexical matching misses thousands of valid concept mappings, which leaves gaps in interoperability between medical data systems.
Key innovation
A "Merged Strategy" that combines lexical and semantic approaches, outperforming either method used alone.
JavaLogMapBioSBERTPythonSPARQLBioPortal API
View code on GitHub ↗References
- Jiménez-Ruiz, E. (2025) LogMap: An Ontology Alignment and Alignment Repair System, Source code [Online]. Available at: https://github.com/ernestojimenezruiz/logmap-matcher (Accessed: 8 December 2025).
- Reimers, N. and Gurevych, I. (2025) Sentence-BERT (Sentence Transformers) Library, Source code [Online]. Available at: https://www.sbert.net/ (Accessed: 8 December 2025).
- LogMap matcher, July 2021 release (dependency). Available at: https://github.com/ernestojimenezruiz/logmap-matcher/releases/tag/logmap-matcher-july-2021