Skip to content

Latest commit

 

History

History
9 lines (8 loc) · 448 Bytes

File metadata and controls

9 lines (8 loc) · 448 Bytes

Questions

How are NER systems usually built?

  • Though rule-based systems exist, typically NER is framed as a supervised learning problem. Take a look at sequence tagging models (HMM, CRF, RNN, LSTM, etc.).
  • For feature-based models, here are few commonly used features:
    • word shape for encoding capitalization (ex. Bob -> Xxx, Virginia -> Xxx)
    • PoS tag
    • prefix and suffix
    • the same features for surrounding words