Inline Annotation

This category encompasses (linguistic) annotation types describing a single structural element. Examples are Part-of-Speech Annotation or Lemmatisation, which often describe a single token.

These annotation types are encoded in an inline fashion in FoLiA, i.e. they appear within the structural element to which they apply (often words/tokens but not necessarily so) and make use of the hierarchical nature of XML.

FoLiA defines the following types of inline annotation:

  • Inline Annotation – This category encompasses (linguistic) annotation types describing a single structural element. Examples are Part-of-Speech Annotation or Lemmatisation, which often describe a single token.

    • Part-of-Speech Annotation – <pos> – Part-of-Speech Annotation, one of the most common types of linguistic annotation. Assigns a lexical class to words.
    • Lemmatisation – <lemma> – Lemma Annotation, one of the most common types of linguistic annotation. Represents the canonical form of a word.
    • Domain/topic Annotation – <domain> – Domain/topic Annotation. A form of inline annotation used to assign a certain domain or topic to a structure element.
    • Sense Annotation – <sense> – Sense Annotation allows to assign a lexical semantic sense to a word.
    • Error Detection Annotation (DEPRECATED) – <errordetection> – This annotation type is deprecated in favour of Observation Annotation and only exists for backward compatibility.
    • Subjectivity Annotation (DEPRECATED) – <subjectivity> – This annotation type is deprecated in favour of Sentiment Annotation and only exists for backward compatibility.
    • Language Annotation – <lang> – Language Annotation simply identifies the language a part of the text is in. Though this information is often part of the metadata, this form is considered an actual annotation.