MALA

The Map of Applications for Linguistic Annotation

MALA MALA is a curated list of applications for collecting and annotating natural language text.

Scope MALA intends to be broad in scope, including apps used by the computer science/NLP community and those used by the linguistics community.

Contributing If you have an application you'd like to add, or you want to submit a correction to an already-listed app, please see our GitHub page.

Classifications

Data Types

• tei: TEI XML
• tags: Annotations which are one-to-one aligned with tokens. Examples include POS tags and interlinear glosses.
• time-aligned: Time alignments between text/annotations and audio/video.
• chunk-spans: Spans (i.e., contiguous sets of tokens), which are NOT allowed to overlap or nest.
• overlapping-spans: Spans (i.e., contiguous sets of tokens), which MAY overlap or nest.
• constituency-tree: Constituency trees, e.g. for PTB-style syntax trees or Rhetorical Structure Theory trees.
• dependency-tree: Dependency trees, e.g. those used for Universal Dependencies treebanks.
• graph: Non-tree graph structures, e.g. for graph-based semantic representations like AMR or UCCA.

Import/Export Formats

• plaintext: A textual representation which might or might not be formatted visually using whitespace, but does not contain any markup.
• proprietary: Some format which is not in wide use outside of the app, and is likely limited just to use within the app.
• csv: A comma- or tab-delimited format.
• xml: Some XML format which may or may not be proprietary.
• json: Some JSON format which may or may not be proprietary.
• conllu: The CoNLL-U format.
• elan-xml: The XML format used by ELAN save files.
• flex-xml: The XML format used in exports by FieldWorks Lexicon Explorer.