Named Entity Recognition (NER) Model

A Named Entity Recognition (NER) Model is a predictive model that a NER system can use to solve an NER task.

• Context:
  • It can be produced by an NER System (e.g. a model-based NER system).
  • ...
• Example(s):
  • http://alias-i.com/lingpipe/demos/models/ne-en-news-muc6.AbstractCharLmRescoringChunker
  • [math]\displaystyle{ f }[/math]("Alexander went to Australia.”) ⇒ “[PERSON|Alexander] went to [LOCATION Australia.]".
  • [math]\displaystyle{ f }[/math]("Alexander went home.”) ⇒ “[PERSON|Alexander] went home.".
  • [math]\displaystyle{ f }[/math]("Microsoft is based in Redmond.”) ⇒ “[ORGANIZATION|Microsoft] is based in [LOCATION|Redmond]."
  • [math]\displaystyle{ f }[/math]("OmpA is one of E. coli’s proteins. It is found in the outer membrane.”) ⇒ “[PROTEIN|OmpA] is one of [ORGANISM|E. coli]’s proteins. It is found in the [LOCATION|outer membrane].” (a Protein NER Task).
  • from Message Understanding Conference.
    • [math]\displaystyle{ f }[/math]("Bruce P. Smith has been named Dean of The Faculty of Science. Dean Smith has served as associate dean for academic affairs since April.”) ⇒ “[PERSON|Bruce P. Smith] has been named [PERSON|Dean]] of [ORGANIZATION|The Faculty of Science]]. [PERSON|Dean Smith] has served as [PERSON|associate dean for academic affairs] since [TIME|April].". (includes an Honorific and a Nested Entity Mention).
  • [math]\displaystyle{ f }[/math]("London is from Toronto") ⇒ “[LOCATION London] is from [LOCATION|Toronto].” A Toponym Recognition Task.
  • …
• Counter-Example(s):
  • a Syntactic Phrase Chunking Model.
  • a Syntactic Parsing Model.
  • a Relation Mention Recognition Model.
  • a Part-of-Speech Tagging Model.
• See: Trained Model.

---

---

References

2011

• http://incubator.apache.org/opennlp/documentation/manual/opennlp.html#tools.namefind
  • The Name Finder can detect named entities and numbers in text. To be able to detect entities the Name Finder needs a model. The model is dependent on the language and entity type it was trained for. The OpenNLP projects offers a number of pre-trained name finder models which are trained on various freely available corpora. They can be downloaded at our model download page. To find names in raw text the text must be segmented into tokens and sentences. A detailed description is given in the sentence detector and tokenizer tutorial. Its important that the tokenization for the training data and the input text is identical.

---