Skip to main content
Identrics
API Framework

Named Entity RecognitionNER API by Identrics

Detect people, organisations, brands and locations across multilingual text.

Talk to us about access
Use it standalone, or compose it with any other Kaspian enrichment - sentiment, NER, topics, summarisation, and more - in a single pipeline.Already identifying the actors behind every narrative cluster in Annex.
Kaspian API
Yesterday …
PersonMerkelOrgSiemensLocationBerlin
… met in Europe.
What the API Does

A single API endpoint, production-ready.

The NER API automatically detects and classifies named entities - people, organisations, brands, locations, products - across unstructured text, turning raw content into structured intelligence you can query, route and analyse.

Powered by an award-winning human-in-the-loop pipeline that combines deep-learning models with expert annotation, the API delivers high precision across long and short form content - in multiple languages.

Common use cases:

  • 01Map every mention of your brand, leadership team and competitors across coverage.
  • 02Power knowledge graphs, entity-based search and recommendation systems.
  • 03Identify the people and organisations driving narratives that put a brand's reputation at risk.
  • 04Support public-sector information integrity work by mapping the actors behind coordinated narratives.
API Capabilities

What you get out of the NER endpoint.

Built and maintained by the Identrics applied AI team. Trained on human-in-the-loop data, benchmarked continuously, and shipped to you as a single, well-documented API.

01

Multilingual extraction

Detect entities across multiple languages with consistent accuracy and no per-language re-integration.

02

Custom entity types

Train the API on your domain - product SKUs, internal org names, legal references - alongside standard entity classes.

03

Disambiguation

Resolve mentions to canonical entities so 'Apple' the company is never confused with 'apple' the fruit.

04

Production-grade throughput

Process millions of documents per day with predictable latency and consistent quality.

Available languages

Models trained and benchmarked across multiple languages out of the box. Additional languages on request.

  • English
  • German
  • French
  • Spanish
  • Bulgarian
  • Italian
  • Chinese (traditional)
  • Portuguese
  • Russian
  • Swedish
  • Dutch
Powers in Production

From algorithmic capability to operational intelligence.

Named Entity Recognition remains available as a standalone Kaspian API. The same capability also supports a focused workflow inside the Identrics product ecosystem.

Annex

Annex uses entity recognition to identify the people, organisations, brands and locations inside narrative clusters, connecting content to the actors shaping it.

Explore Annex
How the API Works

From request to enriched response.

  1. 01

    Send text

    POST raw documents, articles or short-form text to the API - one item or a batch.

  2. 02

    Models extract entities

    Our human-in-the-loop trained models identify and classify every entity in the payload.

  3. 03

    Receive structured JSON

    Get back a clean, typed list of entities with offsets, types and optional canonical IDs.

Where It Fits

Industries already running this API in production.

Brands & Communications

Track mentions of your brand, leadership and competitors across coverage and reviews, and see which actors are shaping narratives that affect your reputation.

Media & Publishing

Tag every article with people, organisations and places automatically. Power related-content recommendations, entity pages and knowledge-graph driven search.

Public Sector & Information Integrity

Map the actors, outlets and organisations behind public narratives so institutions can understand and respond to information disorder with evidence.

NER API

Ready to plug NER into your stack? Tell us the job; we'll spin up a standalone or composed deployment.