Turn raw news into structured, queryable data
Opoint tags every article with standardised topics and disambiguated entities, so 3.5M+ articles a day arrive sorted by what they’re about and who they mention, ready to query, route, and analyse.
Why enrichment is the layer that matters
Raw news isn’t usable at scale.
A feed of 3.5M articles a day is noise until something tells you what each article is about and who it concerns. That structuring is the difference between a stream you can’t search and a dataset you can build on.
Topics & Entities is the layer that does it. Every article arrives tagged: classified by topic against a standard taxonomy, with the people, organisations, and places identified and linked to the identifiers your systems already use. You query and route by meaning, not keywords.
You build the product. Opoint ensures the underlying data is structured, disambiguated, and machine-readable.
Delivered via API or Feed
Our Topics and Entities feature integrates seamlessly into your Opoint news feed, whether you’re using an API or a feed.
This real-time metadata transforms your approach to news data analysis.
Dive into Topics
With our integration of the International Press Telecommunications Council (IPTC) Media Topics, you get a comprehensive taxonomy of over 1000 terms designed for media content categorisation.
We can also build topic-based feeds for more targeted analysis.
Identify the key players, and link them to your data
Opoint scans each article to identify the people, organisations, and locations involved, disambiguates them, and links each entity to the identifiers your systems already use. A competitor or counterparty surfaces across name variants, subsidiaries, and languages, and joins cleanly to the reference data you already hold.
Product Details
Topics
Every article is classified against the IPTC Media Topics taxonomy: 1000+ standardised terms, the global newsroom standard. Each tag carries a label, the IPTC Media Topic ID, and a relevance score, so you can filter by subject and confidence, not guesswork.
Topic tags include:
- Label – the name of the IPTC Media Topic
- Media Topic ID – the unique IPTC Media Code
- Score – the overall confidence that the article revolves around the topic.
Entities
Opoint identifies people, organisations, and locations in every article, disambiguates them (so “Apple” the company is never confused with the fruit, and two people sharing a name stay distinct), and links each entity to the identifiers your systems already run on:
Entity tags include:
- Entity type – person, location, or organisation.
- Entity – name of person, location, or organisation.
- Wikidata ID – a link to additional information about the entity (when possible).
- FIGIs – Financial Instrument Global Identifiers to uniquely identify financial instruments.
- LEIs (Legal Entity Identifiers) are used to identify legal entities globally.
- Crunchbase ID – a link to a Crunchbase profile for businesses.
Subscribe to our newsletter
FAQ
What are topics and entities in news enrichment?
Topics classify what an article is about, using the IPTC Media Topics taxonomy of 1000+ standardised terms. Entities identify who and where they are mentioned: the people, organisations, and locations, disambiguated and linked to standard identifiers. Together, they turn unstructured news into queryable, structured data.
Which entity identifiers does Opoint support?
Entities are linked to Wikidata, LEI (Legal Entity Identifier), FIGI (Financial Instrument Global Identifier), and Crunchbase, so news data joins cleanly to the reference data your systems already use.
How is this delivered?
As structured metadata attached to each article, delivered by API alongside the news feed. It integrates into your existing pipeline rather than requiring a separate tool.
Does Opoint provide the analysis or the data?
Opoint provides the structured, enriched data. You run the analysis, screening, or product on top of it. The enrichment layer makes that work possible; it doesn’t replace it.
Can entities be tracked across languages?
Yes. Disambiguation and identifier linking work across Opoint’s multilingual coverage, so an organisation surfaces consistently whether it’s reported in English or in another language.
Related
