Naming operations

Naming operations can be divided into passive ones, applied for checking purpose, and active ones, for naming a new entity. While both can be plugged into naming transformation patterns, only passive operations can be used in naming detection patterns.

Currently, the list of passive naming operations is as it follows:

• delimiter_xxx(?A): checks whether entity name uses given delimiter, e.g. underscore, camel-case or hyphen where xxx stands for the delimiter,
• verb_form(?A): checks whether noun $?A$ has a correspoding verb form. Currently, it is implemented using WordNet and its semantic information ``derivationally related forms'', where we check whether particular related form is a verb. We use Stanford part-of-speech tagger: http://nlp.stanford.edu/software/tagger.shtml. It stops when it finds the first verb form.
• head_noun(?A): returns a term on which all other tokens, from the whole phrase, are dependent. Similarly, there is a head_term(?p) for properties with a little bit different detection rules,
• complement_head_noun(?A): returns a complement of the head_noun to the union of all tokens appearing in entity name,
• passive_verb(?A): returns a passive form of verb from entity name of ?A

Currently, the list of active naming operations is as it follows:

• make_delimiter_xxx(?A): changes delimiters in ?A to underscore, camel-case or hyphen where xxx stands for new delimiter.
• make_passive_verb(?A): returns a passive form of given verb in the case of a verb. In the case of a noun, it returns a passive form of a correspoding verb form to given noun, see explanation for passive naming function verb_form() above. If there is no such form it returns string on input.

The list of implemented naming operations is being gradually extended as needed for supported transformation patterns. In Table 1, there are all so far considered naming operations regardless their availability in transformation patterns as such.

Table 1: Naming operations

naming operation	output	note
tokenize(entity)	token[]	implemented in PatOMat core
head_noun(class|individual)	string	implemented in PatOMat core
head_term(property)	token	implemented in PatOMat core
object_term(property)	token	implemented in PatOMat core
delimiter(entity)	delimiters where delimiters::= "underscore"$\vert$"hyphen"$\vert$"camel_case"	implemented within tokenize(entity)
make_delimiter_xxx(entity) where xxx stands for particular delimiter	string	implemented in PatOMat Ontology Transformation (OT)
singular(entity)	boolean	implemented within PatOMat NLP
plural(entity)	boolean	implemented within PatOMat NLP
pos(entity):part_of_speech	part_of_speech	implemented within PatOMat NLP
passive(verb)	boolean	implemented within PatOMat NLP
make_passive_verb(entity)	token	implemented in PatOMat OT
multipleTokenName(entity)	boolean	implemented in PatOMat core
hyperonym(token,token)	boolean	implemented in PatOMat core using WordNet
similar(token,token)	boolean	implemented in PatOMat core using Roget
verb_form(token)	boolean	implemented in PatOMat OT
complement_head_noun(token)	token[]	not implemented yet