Output Description File - VincTheSecond/rextractor GitHub Wiki

XML document describes entities and relations detected in the input document. Each entity is defined as list of text-chunks (element annotation in HTML document) and each relation is defined as a set of entities. Here is description of elements used in XML document:

• document

  • Root element.
  • No attributes.

• document → metadata

  • Document metadata section. So far undefined.
  • No attributes.

• document → entities

  • List of entities detected in the input document.
  • No attributes.

• document → entities → entity

  • Description of one entity detected in the document.
  • Attributes:
    • entity_id
      • Unique id of the entity. This is is used for entity addressing in relation definitions.
    • dbe_id
      • Identifier of entry in the Database of Entities. Defined if entity was detected in Entity component.
    • chunk_ids
      • List of chunks (element in HTML document) which creates the entity.

• document → entities → entity → dependency_tree

  • Definition of the dependency tree for the entity. Each token of the entity has defined its node in the tree. Each node (element ) is descripted by several attributes:
    • form
      • Original form of the token.
    • lemma
      • Base form of the token.
    • ord
      • Ord number of the token in the entity.
    • parent
      • Ord number of the parent token.

• document → relations

  • List of detected relations
  • No attributes.

• document → relations → relation

  • Description of one specific relation.
  • Attributes:
    • relation_id
      • Unique relation identifier
    • dbr_id
      • Id of the query which describes relation and its RDF representation.
    • (subject|predicate|object)_ids
      • List of entity ids which are on the position of the $1.
    • (subject|predicate|object)_concept
      • Concept of the entity used in RDF transformation.