README.md

NER Benchmark

Named Entity Recognition benchmark dataset.

Entities

• PER (Person)
• ORG (Organization)
• LOC (Location)
• MISC (Miscellaneous)

Task

Predict entity labels for each token.

Metrics

• Precision
• Recall
• F1 (entity-level)

Format

{
  "tokens": ["Apple", "is", "great"],
  "ner_tags": ["ORG", "O", "O"]
}