Skip to content

SYNCON attribute

With syncon ID

The SYNCON attribute identifies a concept by specifying the numeric ID of a syncon contained in the knowledge graph. A token will be recognized in a text, if, during the disambiguation process, it is associated with the specified syncon.

The syntax is:

SYNCON(ID1[, ID2, ...])

where:

  • SYNCON is the attribute name and must be written in uppercase.
  • ID# is the identification number of a concept (syncon) contained in the knowledge graph.

Since a syncon can be made up of one or more lemmas representing the same concept, all synonyms, variants, abbreviations etc. that are part of the same syncon will be matched, if a corresponding text is found. Synonyms and variants are also recognized both in their base form and inflected forms.

When the SYNCON attribute is used in a rule, a token will be identified in a document, only if it is analyzed (disambiguated) as being an instance of the syncon specified in the rule. The token in the document must match one of the lemmas that are part of the syncon. However this is not a sufficient condition for a token to match a SYNCON rule. In fact, many lemmas contained in the knowledge graph are polysemic, meaning they have several meanings or they represent different concepts. For a SYNCON rule to be verified, the disambiguator must associate the token in the document to the meaning represented by the syncon in the rules. This means that the SYNCON attribute not only considers the form of a word, but also its contextual meaning.

For example:

SYNCON(17200)

The above refers to the concept of house with a meaning of a building for human habitation. In the knowledge graph, this syncon contains three lemmas: house, casa and the abbreviation ho.. If a rule uses this syncon ID, every time that house or casa or ho. are found in a document with a meaning of a dwelling that serves as living quarters for one or more families, the above rule will be verified/triggered.

Let's compare the statement above with the following:

SYNCON(123085)

This syncon also contains the lemma house, but instead of referring to a building for human habitation, it refers to house as a company that designs, produces and sells clothing and accessories. In this case, the lemma house is found in the syncon together with the lemma fashion house. As in the case above, a rule using this syncon ID (123085) will be verified/triggered every time that any of the above lemmas (house, fashion house) are found in a document and their contextual meaning is the one represented by the selected syncon.

Using the SYNCON attribute, it is also possible to select a list of several IDs:

SYNCON(12985, 15039, 15743)

Such a rule would recognize the concepts apartment, country house and duplex in a document and the rule would search for all of the synonyms and variants grouped in each syncon.

Pay careful attention however when using the SYNCON attribute with terms that are difficult to disambiguate or with terms that contain unwanted synonyms. In this case, it is advisable to furtherly specify the rule using an instruction of LEMMA or KEYWORD.

With UNKNOWN

It is also possible to use the attribute SYNCON with the value UNKNOWN instead of a syncon ID, as shown below.

SYNCON(UNKNOWN)

Using this syntax, only "unknown" elements will be matched in a text. Unknown elements include words which are not contained in the knowledge graph and unknown entities which have been assigned a virtual supernomen by the disambiguator.

Warning

The value UNKNOWN, if used alone, can be extremely powerful and hyper generative. In fact, it is designed and advised to be used in combination with other attributes.