Recommender API Schema - ExtractResponseBody

Abstract

Schemata for the Recommender API Calls

The response from the extract endpoint containing all concepts found on the basis of the input text and parameters.

message*[...]
extractedConcepts[...]
    [ScoredConcept]
        [MatchingLabelModel]
            [MatchedText]
                [TokenPosition]
shadowConcepts[...]
    [ShadowConcept]
        [ShadowTerm]

Note

An asterisk (*) next to a parameter indicates a mandatory parameter.

The response from the extract endpoint containing all concepts found on the basis of the input text and parameters.

Parameter	Type	Description	Details
`message*`	string	Short descriptive message with the results of the extraction.	response codes: 200 for successful response, 400 for bad request, 5XX - for an unexpected internal error.
`extractedConcepts`	array/list	List of concepts extracted from the input text.	Contains `ScoredConcept` where the matchingLabels parameter uses `MatchingLabelModel` in which the `matchedTexts` parameter contains `MatchedText` in where the `positions` using the `TokenPosition` are contained.
`shadowConcepts`	array/list	List of shadow concepts matching the input text.	Contains `ShadowConcept` where `shadowConceptTerms` are used containing `ShadowTerm`.

Parameter

Type

Description

Details

message*

string

Short descriptive message with the results of the extraction.

response codes:

200 for successful response, 400 for bad request, 5XX - for an unexpected internal error.

extractedConcepts

array/list

List of concepts extracted from the input text.

Contains ScoredConcept where the matchingLabels parameter uses MatchingLabelModel in which the matchedTexts parameter contains MatchedText in where the positions using the TokenPosition are contained.

shadowConcepts

array/list

List of shadow concepts matching the input text.

Contains ShadowConcept where shadowConceptTerms are used containing ShadowTerm.

A scored concept; used for instance in the extractedConcepts parameter.

Parameter	Type	Description	Details
`uri*`	string	URI of the concept.	example: http://www.w3.org/TR/2004/REC-owl-guide-20040210/wine#Merlot
`label*`	string	Preferred label of the concept.	example: Merlot
`frequency*`	integer	Number of occurrences in the input text.	example: 43
`score*`	number	Extraction score.	maximum: 100 `exclusiveMaximum`: false minimum: 0 `exclusiveMinimum`: false example: 55.3
`corpusScore`	number	Corpus score; only available with corpus scoring enabled.	maximum: 100 `exclusiveMaximum`: false minimum: 0 `exclusiveMinimum`: false example: 18
`matchingLabels`	list of MatchingLabel objects	Matching information for the label of this concept related to the input text.	contains `MatchingLabelModel`, which contains `MatchedText` where `TokenPosition` is contained.

Matching information related to the labels of this concept in relation to the input text.

Parameter	Type	Description	Details
`label`	string	Found label.	example: Merlot
`predicate`	string	Predicate name of the given label.	example: prefLabel
`language`	string	Language of the label.	`maxLength`: 2 `minLength`: 2 example: en
`frequency`	integer	Number of occurrences of the label in a given text.	example: 3
`matchedTexts`	list of MatchedText objects	Text fragments matching the label and its position.	contains `MatchedText`, which then contains `TokenPosition`

Text fragments matching the label along with their position.

Parameter	Type	Description	Details
`matchedText`	string	Text which was matched.	example: fox
`frequency`	integer	Frequency how often the particular text fragment was matched in the text.	example: 3
`positions`	list of TokenPosition objects	All positions where this fragment of text was matched in the input text.	contains `TokenPosition`

Parameter

Type

Description

Details

matchedText

string

Text which was matched.

example:

fox

frequency

integer

Frequency how often the particular text fragment was matched in the text.

example:

positions

list of TokenPosition objects

All positions where this fragment of text was matched in the input text.

contains TokenPosition

All positions where the particular fragment of text was matched in the input text.

Parameter	Type	Description	Details
`beginIndex`	integer	Begin index of the particular match.	example: 15
`endIndex`	integer	End index of the particular match.	example: 18

Parameter

Type

Description

Details

beginIndex

integer

Begin index of the particular match.

example:

endIndex

integer

End index of the particular match.

example:

shadowConcepts is a list of shadow concepts matching the input text. The table below lists the individual parameters of a ShadowConcept where a ShadowConcept is defined as a concept not explicitly defined in the input which is however often associated with the input topics when compared to the corpus.

Parameter	Type	Description	Details
`uri`	string	URI of the concept.	example: http://www.w3.org/TR/2004/REC-owl-guide-20040210/wine#Merlot
`label`	string	Preferred label of the concept.	example: Merlot
`score`	number	Shadow concept score.	maximum: 100 `exclusiveMaximum`: false minimum: 0 `exclusiveMinimum`: false example: 0.4
`shadowConceptTerms`	list of ShadowTerm objects	Terms extracted from the input text contributing to the calculation of the shadow concept.	contains `ShadowTerm`

A term extracted from the input text and related to a shadow concept.

Parameter	Type	Description	Details
`textValue*`	string	Term from the input text related to the shadow concept.	example: fox
`score*`	number	Relevance of this term in relation to the shadow concept.	maximum: 100 `exclusiveMaximum`: false minimum: 0 `exclusiveMinimum`: false example: 23

Parameter

Type

Description

Details

textValue*

string

Term from the input text related to the shadow concept.

example:

fox

score*

number

Relevance of this term in relation to the shadow concept.

maximum: 100

exclusiveMaximum: false

minimum: 0

exclusiveMinimum: false

example:

In this section: