Organizations

Trendtracker treats organizations as first-class entities: companies, research institutions, and other legal or operational units that show up in investments, patents, news, and other signals. They are not just labels in text—they are stable records in a knowledge graph that we connect to the rest of the platform.

A graph built on trusted registries

The graph is grounded in large, public reference databases that define who an organization is in the real world. We integrate and reconcile identifiers and attributes from sources such as:

• GLEIF (Global Legal Entity Identifier Foundation) — legal entity identifiers and corporate structure where available.
• Wikidata — rich linked open data and cross-references to many other systems.
• Research Organization Registry (ROR) — persistent identifiers and metadata for research-performing organizations.

Those foundations give us shared IDs, hierarchies, and aliases so that “the same organization” can be recognized consistently across documents, filings, and datasets—even when the surface text differs (subsidiary names, abbreviations, rebrands).

Linking your data to the graph: Entity resolution

The content Trendtracker ingests and processes (articles, announcements, regulatory text, and more) rarely arrives with a clean, canonical org ID. We run entity resolution: matching mentions and records in that content to the right graph nodes—using names, context, known relationships, and confidence from the knowledge we have already ingested.

Once resolved, activity and attributes can be linked together: the same organization can be seen across topics, time, and indicator types without fragmenting into duplicate or ambiguous strings. That is what makes organization-centric analysis—who is active in a space, who is investing, who appears in a deal—composable with the rest of Connect (topics, indicators, and related concepts).