IBM Information Server Governance 11.7 features Machine Learning capabilities which intelligently reduces manual intensive processes of Term assignments to technical metadata in Information Governance Catalog and Information Analyzer.
We use Supervised machine learning model to solve this problem. This model is initially trained with your business glossary and terms in the Information Governance Catalog product. As soon as you import your technical metadata assets to the tool, these models are ready to detect related terms for the technical assets and present users matches with confidence score. A 100% matched term will automatically be assigned, but others are presented as options. User’s selection will be automatically captured and provided to the model as feedback. This in turns triggers retraining of the model for better outcome in next user action.