Linguistic Models
A natural language consists of two parts:
- lexicon: vocabulary of the language
- grammar: set of rules to combine words from the lexicon in a meaningful way
Shallow Model
Machine Learning techniques in NLP are usually shallow
- don’t care much about grammar
Shallow Parsing:
- JMLR Special Issue on Shallow Parsing [http://www.jmlr.org/papers/special/shallow_parsing02.html]
- Machine Learning Approaches to Shallow Parsing [http://www.jmlr.org/papers/v2/hammerton02a.html]
Deep Model
- Deep linguistic processing is a natural language processing framework which draws on theoretical and descriptive linguistics.
- It models language predominantly by way of theoretical syntactic/semantic theory
Sources
- http://en.wikipedia.org/wiki/Deep_linguistic_processing