Glossary · 7 · Product information systems and data roles
Semantic model (BI)
Also known as: Semantic layer, BI semantic layer, Metrics layer
In business intelligence, a semantic model is the layer between data sources and reports that defines tables, relationships, calculations (measures) and business names, so that dashboards and queries use the same definitions of each figure.
- Data management
- Data analytics
In one sentence
In business intelligence, the layer that defines tables, relationships, measures and business names over data for reports and dashboards.
Example
The service dashboard counts “open documentation defects” through a measure in the semantic model; changing that measure changes the figure on every report that uses it.
Explained in context
Context cards connect this term with others to answer one question. Also in British English and German.
How it applies
- Figures are derived data: Every figure on a dashboard is computed by the semantic model from source data. When a documentation team reports figures — open defects, translation status, coverage — the definition in the model decides what the figure means.
- Shared definitions: A semantic model helps when the same figure must mean the same thing on every report; it does not make the underlying values released or correct.
Semantic model (BI) vs. semantic model (information modeling)
A semantic model in information modeling describes the concepts, relations and meaning of a domain, often formally (ontologies, RDF). The BI semantic model describes how data is aggregated and named for reports. The two share a name, not a purpose: the first defines what things are, the second how figures are calculated.
External references
- Semantic layer (overview): https://en.wikipedia.org/wiki/Semantic_layer