Named Entity Recognition (NER) Explorer Tool

Combine common Named Entity Recognition (NER) categories with your own custom entity types! This tool uses both traditional NER models and GLiNER for comprehensive entity extraction, allowing you to explore what NER looks like in practice.

How to use this tool:

  1. 📝 Enter your text in the text area below
  2. 🎯 Select a model from the dropdown menu
  3. ☑️ Select common entity types you want to identify (PERSON, ORGANIZATION, LOCATION, etc.)
  4. ✨ Add custom entities (comma-separated) like "relationships, occupations, skills"
  5. ⚙️ Adjust confidence threshold
  6. 🔍 Click "Analyse Text" to see results (NB: common/custom entities which overlap are shown with split-colour highlighting)
  7. 🔄 Refresh the page to try again with new text
💡 Top tip: No model is perfect - all can miss and/or incorrectly identify entity types.
0.1 0.9
ℹ️ Understanding the Confidence Threshold

The confidence threshold controls how certain the model needs to be before identifying an entity:

  • Lower values (0.1-0.3): More entities detected, but may include false positives
  • Medium values (0.4-0.6): Balanced detection with moderate confidence
  • Higher values (0.7-0.9): Only highly confident entities detected, may miss some valid entities

Top Tip: Start with 0.3 for comprehensive detection, then adjust based on your needs.

🎯 Common Entity Types
Select NER Model for Common Entities
Select Common Entities

✨ Custom Entity Types - Powered by GLiNER

Examples:

  • relationships, occupations, skills
  • emotions, actions, objects
  • medical conditions, treatments
ℹ️ Entity Type Definitions
Date (DATE):
Absolute or relative dates or periods
Event (EVENT):
Named hurricanes, battles, wars, sports events, etc.
Facility (FAC):
Buildings, airports, highways, bridges, etc.
Geopolitical Entity (GPE):
Countries, cities, states
Language (LANG):
Any named language
Location (LOC):
Non-GPE locations - Mountain ranges, bodies of water
Miscellaneous (MISC):
Entities that don't fit elsewhere
Nationalities/Groups (NORP):
Nationalities or religious or political groups
Organization (ORG):
Companies, agencies, institutions, etc.
Person (PER):
People, including fictional characters
Product (PRODUCT):
Objects, vehicles, foods, etc. (Not services)
Work of Art (Work of Art):
Titles of books, songs, movies, paintings, etc.

💡 No example text to test? No problem!

Simply click on one of the examples provided below, and the fields will be populated for you.
Examples
📝 Text to Analyse Select Common Entities Custom Entities (comma-separated) 🎚️ Confidence Threshold Select NER Model for Common Entities

📚 Model Information & Documentation (incl. details on usage terms)

Learn more about the models used in this tool:


This NER Explorer Tool was created as part of the Digital Scholarship at Oxford (DiSc) funded research project:
Extracting Keywords from Crowdsourced Collections.



The code for this tool was built with the aid of Claude Opus 4.