Sparv Concepts

This section provides an overview of the key concepts and components of Sparv, including the corpus directory structure, configuration files, annotations, annotators, modules, and plugins. Some of these concepts are explained in more detail in other sections of the user manual.

Corpus Directory

Each corpus that you want to process with Sparv needs to be stored in a separate directory. This directory is referred to as a corpus directory. A corpus directory contains a corpus configuration file (config.yaml) and a directory with source files. The source files are the documents to be annotated, in a format that Sparv can process, such as XML or plain text.

The corpus directory structure and source file requirements are described in more detail in the Preparing Your Corpus section.

Configuration File

The configuration file (config.yaml) specifies the desired annotations, export formats, and other processing instructions for Sparv. The existence of a configuration file is what makes a directory a corpus directory, and most Sparv commands won't work without it.

The configuration file is written in YAML format. For a detailed guide on creating corpus configuration files, refer to the Corpus Configuration section.

Modules and Plugins

Modules are the building blocks of Sparv, each responsible for a specific part of the annotation process. Modules contain one or more processors, of which there are several types: importers for reading source files, annotators for analyzing the text and adding annotations, exporters for saving the results, and more.

Sparv comes with a set of built-in modules that cover the entire workflow from importing text data, annotating it with linguistic features, exporting the results in different formats, to deploying the annotated corpus.

Plugins are external modules that extend Sparv's functionality. They can be used to add support for additional third-party tools, custom annotations, or other features that are not part of the Sparv core. Writing plugins is covered in the Developer's Guide.

Annotations and Annotators

Annotations are the linguistic features that Sparv adds to the text. Annotators are a type of processor that adds annotations to the text. Examples of annotations include part-of-speech tags, lemmas, named entities, and syntactic structures.

How to specify which annotations to add to your corpus is described in the Corpus Configuration section. For a list of available annotations and annotators, run the sparv modules command, as described in the Running Sparv section. You can also find a list of available annotations in the Available Analyses section in the documentation.