Dictionary-based Named Entity Recognition Task
(Redirected from Dictionary-based NER Task)
Jump to navigation
Jump to search
A Dictionary-based Named Entity Recognition Task is a heuristic NER task that makes use of a lexical database.
- Context:
- It can be solved by a Dictionary-based NER System.
- …
- Example(s):
- a Protein Recognition Task, based on a Protein Dictionary.
- …
- Counter-Example(s):
- See: Dictionary-based NER Algorithm, Dictionary-based NER System.
References
2010
- http://alias-i.com/lingpipe/demos/tutorial/ne/read-me.html
- QUOTE: In many applications, it is relatively straightforward to compile a list of names (and aliases) for entities and their types. For instance, westlaw.com has lists of all the registered lawyers in the United States. Such lists are extremely helpful for finding mentions of lawyers in legal documents such as case law or court transcripts. A baseball site such as mlb.com has a list of all current and retired baseball players to work from. … The class dict.ApproxDictionaryChunker implements a dictionary-based chunker that performs approximate matching. An approximate dictionary chunker is populated with a dictionary, the same way as an exact dictionary chunker. But when it is run, it doesn't just look for exact matches, but for all matches within a fixed weighted edit distance threshold.