Dealing with knowledge evolution and its consequences on terminologies, mappings, and data is a big challenge for the industry. Current surveys show that 10% of terminologies representing domain knowledge evolve every year, impacting on the quality of companies’ assets such as knowledge bases, decision-support systems, search engines, etc. In extreme cases, without any maintenance, by 2025 more than 35% of terminology-dependent artifacts will become obsolete.
Dynaccurate was developed by Luxembourg Institute of Science and Technology to support industries dealing with issues raised by terminology evolution and its effect on Semantic Web data. The tool is the only existing solution that implements advanced artificial intelligence techniques to (semi-)automatically repair outdated mappings and semantic annotations.
Dynaccurate can identify what has changed in the terminologies, assist manual curation of mappings and semantic annotations, and provide machine readable information that can be used for further automation opportunities.
Evaluated with biomedical and legal multilingual datasets, Dynaccurate reached a very good precision for repairing mappings and semantic annotations.