Zubnet AILearnWiki › Named Entity Recognition
Using AI

Named Entity Recognition

NER, Entity Extraction
Identifying and categorizing named entities in text — people, organizations, locations, dates, monetary amounts, and other proper nouns. In "Apple announced a $3B investment in Munich on Tuesday," NER identifies Apple (Organization), $3B (Money), Munich (Location), and Tuesday (Date). It's a foundational NLP task used in information extraction, search, and knowledge graph construction.

Why it matters

NER is the backbone of structured information extraction from unstructured text. Every search engine, news aggregator, and intelligence system uses NER to understand what a document is about. It's also the first step in building knowledge graphs from text — you can't build relationships between entities you haven't identified.

Deep Dive

NER is typically framed as a sequence labeling task: each token gets a label like B-PER (beginning of person name), I-PER (inside person name), O (not an entity). The BIO tagging scheme handles multi-word entities: "New" gets B-LOC, "York" gets I-LOC. Fine-tuned BERT models are the standard for high-accuracy NER, though spaCy's built-in NER is popular for quick, good-enough extraction.

Domain-Specific NER

General NER models handle common entity types (person, org, location, date). Domain-specific applications need custom types: medical NER extracts drugs, symptoms, and dosages. Legal NER extracts case numbers, statutes, and parties. Financial NER extracts ticker symbols, financial metrics, and regulatory references. These require domain-specific training data, which is expensive to annotate but dramatically improves extraction quality in specialized contexts.

NER with LLMs

LLMs can perform NER through prompting: "Extract all person names and organizations from this text and return as JSON." This is slower and more expensive than dedicated NER models but handles novel entity types without training data and works across languages out of the box. For production systems processing millions of documents, dedicated models win on cost. For ad-hoc extraction or uncommon entity types, LLMs win on flexibility.

Related Concepts

← All Terms
ESC