package textannotation

Get desktop application:
View/edit binary Protocol Buffers messages

Constituent message structure

Used in: ViewDataProto

• required string label = 1

  Label of the constituent

• optional double score = 2

  Score of the constituent

• required int32 start = 3

  Start index of the span Span indices correspond to indices in the tokens list.

• required int32 end = 4

  End index of the span + 1 Span indices correspond to indices in the tokens list. Usage: a span corresponds to [start, end) - end is not inclusive.

• map<string, string> properties = 5

  Attribute information

• map<string, double> labelScoreMap = 6

  Map containing scores to each label

Relation message structure

Used in: ViewDataProto

• required string relationName = 1

  Name of the relation

• optional double score = 2

  Score of the relation

• required int32 srcConstituent = 3

  Identifier for the source constituent

• required int32 targetConstituent = 4

  Identifier for the target constituent

• map<string, string> properties = 5

  Attribute information

• map<string, double> labelScoreMap = 6

  Map containing scores to each label

Sentence message structure

Used in: TextAnnotationProto

• required string generator = 1

  Class name for the generator

• optional double score = 2

  Score of the sentence

• repeated int32 sentenceEndPositions = 3

  List of sentence end positions

Root TextAnnotation message

• required string corpusId = 1

  Corpus Identifier

• required string id = 2

  Document Identifier

• required string text = 3

  Raw text corresponding to this annotation

• repeated string tokens = 4

  List of tokens

• repeated TokenOffsetsProto tokenOffsets = 5

  Token Offset information

• required SentenceProto sentences = 6

  List of Sentences

• repeated ViewProto views = 7

  List of Views

• map<string, string> properties = 8

  Attribute information

Token Offset message structure

Used in: TextAnnotationProto

• required string form = 1

  Surface form of the token

• required int32 startCharOffset = 2

  Offset of the start character in source text

• required int32 endCharOffset = 3

  Offset of the last character in text + 1 Note: End offset is not part of the current token. [startCharOffset, endCharOffset)

ViewData message structure

Used in: ViewProto

• required string viewType = 1

  Type of the view. Corresponds to class name in core-utilities

• required string viewName = 2

  Name of the view

• required string generator = 3

  Class name of the generator

• optional double score = 4

  Score of the View

• repeated ConstituentProto constituents = 5

  List of constituents in the view Note: Order of constituents stored is important and should be preserved

• repeated RelationProto relations = 6

  List of relations in the view Note: Order of constituents stored is important and should be preserved

View message structure

Used in: TextAnnotationProto

• required string viewName = 1

  Name of the view

• repeated ViewDataProto viewData = 2

  View Data information