NLP- Specific Operators

A group of operators is specifically designed to be useful in NLP models. They are

ACTIVECHILD
Calls diffuse operator.
ADDPROPERTY
Adds property to node or changes property
ALTERNATIVES
Allows alternatives to be specified for words or objects
ANDPARSE
Maintains consistency in parse structure
ANDWITH
Requires two alternatives to be present for propagation
BRIDGE
Sequences the operations during parse structure building
CHECKLEFT
Checks by running along chain looking for things
COLLOCATION
Allows for specification of collocations in various ways
INSERTMARKER
Inserts symbol into parse chain
ISCONSISTENT
Introduces new property to an object based on its context
MEANING1
Supports different meanings
MEMBERGROUP
Connects to members of ObjectGroup
NOCHILD
Prevents finding particular parent
NOTCONSISTENT
Prunes inconsistent meanings
NOTSYMBOL
Any symbol but this one.
NOTWITH
Requires other specific alternatives not to be present for propagation to occur
OBJECTGROUP
A representation of a group of objects
OBJECTONLY
Gathering object properties while ignoring word properties
OBJECTTEST
Embedding an operator which tests the objects in an ObjectGroup
PARSE
Joins pairs of words in parse chain at foot of parse structure
SIMILARMAP
Using related objects to improve anaphoric discrimination
STRUCTURE1
Builds parse structure by cloning itself when a pattern match is detected
SYNONYMMAP
Linking synonyms that require mapping of parameters, such as ToBuy and ToSell
WITHPROPERTY
Checks real properties of objects in grammatical structure

Operators that form a class hierarchy are also used extensively in NLP models.

Map Functions perform constraint reasoning in maps.

See Operators in large parsing model

NLP