Glossary

This wiki has been moved to https://dev.languagetool.org - this page is archived and is not updated anymore

After the Deadline

After the Deadline is open-source language-checking software. Refer to http://open.afterthedeadline.com.

API

An API (application programming interface) is a specification that describes how different software components interact. Refer to http://foldoc.org/API.

AtD

AtD is an abbreviation of After the Deadline.

attribute

In an XML document, an attribute is a part of an element. An attribute gives additional information about an element or the content of an element.

automaton

Refer to finite-state automaton.

AWK

AWK is a programming language that is designed for text processing. Refer to http://en.wikipedia.org/wiki/AWK.

base form

The term base form is a synonym of the term lemma.

bilingual text

A bilingual text is a text that contains 2 languages. Segments (usually sentences) are 'aligned'. For example, in a spreadsheet, one column contains source sentences and another column contains translated sentences. Refer to http://wiki.languagetool.org/checking-translations-bilingual-texts.

bitext

Refer to bilingual text.

Brill tagger

A Brill tagger is a type of POS tagger. A Brill tagger uses a small set of language rules to assign a part of speech to a word. Refer to http://en.wikipedia.org/wiki/Brill_tagger and http://acl.ldc.upenn.edu/H/H92/H92-1022.pdf.

category

In the grammar.xml file, a category is an element that is used to put rules into groups. Refer to http://wiki.languagetool.org/development-overview.

chunk

A chunk (or phrase) is a group of one or more words that has a particular part of speech. Refer to http://wiki.languagetool.org/using-chunks. Refer also to noun chunk; verb chunk.

chunk tag

A chunk tag is a tag that specifies the part of speech that a chunk has. A chunk tag also gives other information such as whether a word in a chunk is at the start of the chunk. Refer to http://wiki.languagetool.org/using-chunks#toc1.

chunker

A chunker is software that partitions plain text into sequences of semantically related words (http://cogcomp.cs.illinois.edu/page/software_view/13). While a POS tagger only works on single words, the chunker uses the results of the POS tagger and covers sequences longer than one word.

committer

A committer is a person who can put changes to LanguageTool on the LanguageTool repository in GitHub. The changes can be to the software, the rules, or to both. Refer also to language maintainer.

configuration file

A configuration file is a file that stores a user's settings or preferences. In LanguageTool, the configuration file is .languagetool.cfg in the user's home directory.

Constraint Grammar

Constraint Grammar is a language-independent method of parsing text. Refer to http://www.ling.helsinki.fi/~fkarlsso/CG-book_1.pdf.

corpus query language

A corpus query language is a method for extracting text samples from a corpus of text. [http://www.cl.uzh.ch/studies/theses/lic-master-theses/lizcharlottemerz.pdf]

curl

curl is command-line software that uses URL syntax to get or to send files. Refer to http://curl.haxx.se/docs/faq.html#What_is_cURL.

disambiguation.xml

disambiguation.xml is the LanguageTool file that contains the rules for a disambiguator. Not all languages have a disambiguation.xml file. Refer to http://wiki.languagetool.org/developing-a-disambiguator.

disambiguator

A disambiguator is software that tries to determine the part of speech that a word has in a particular context. (In many languages, a word can have more than one part of speech. For example, in English, the word help is both a noun and a verb. In the sentence, "Give me help," the word help is a noun.) In LanguageTool, if a language has a rule-based disambiguator, then the rules are in the disambiguation.xml file.

Eclipse

Eclipse is software that helps software developers to write programs in the Java programming language. Refer to http://www.eclipse.org.

element

An element is part of an XML document. Each XML document contains a hierarchy of elements. Each element represents the structure of some information. For example, in the grammar.xml file, each rule element contains information about a particular grammar rule. An element can have one or more attributes.

false friend

False friends are a pair of words or phrases in 2 different languages that look or sound almost the same, but which have different meanings. For example, the English word gift means poison in German.

filter (verb)

In the context of disambiguation, to filter means to remove all but one specified POS tag. Refer to http://wiki.languagetool.org/developing-a-disambiguator#toc4.

finite-state automaton (FSA)

A finite-state automaton is a device that can be in one of many states. In certain conditions, the FSA can change to a different state. Refer to http://en.wikipedia.org/wiki/Finite-state_machine and to http://csunplugged.org/finite-state-automata.

finite-state machine (FSM)

Refer to finite-state automaton.

FSA

Refer to finite-state automaton.

GitHub

GitHub is a website that helps software developers to collaborate, to review software code, and to manage software code. The software code for LanguageTool is on https://github.com/languagetool-org/languagetool/.

grammar.xml

grammar.xml is the LanguageTool file that contains the error detection rules that LanguageTool uses in its evaluation of text. Refer to http://wiki.languagetool.org/development-overview.

GUI

A GUI is a 'graphical user interface'. Typically, a screen shows small images, boxes in which to enter text, and buttons to click. Refer to http://en.wikipedia.org/wiki/Graphical_user_interface.

Hunspell

Hunspell is a spell checker and a morphological analyzer. Hunspell is designed for languages that have rich morphology, complex compounds, and complex character encoding. Refer to http://sourceforge.net/projects/hunspell/ and to http://hunspell.sourceforge.net.

immunize

To immunize means to prevent LanguageTool from matching one or more words. Refer to http://wiki.languagetool.org/developing-a-disambiguator#toc9.

inflected (adj)

Refer to inflection.

inflection

An inflection is the form of a particular lemma. For example, In English, the lemma break has the inflections break, breaks, broke, broken, and breaking.

Internationalization Tag Set (ITS)

The Internationalization Tag Set (ITS) is a technology that helps people to create XML that is internationalized and that can be localized effectively. Refer to http://www.w3.org/TR/2007/REC-its-20070403/ and to http://www.w3.org/International/multilingualweb/lt/drafts/its20/its20.html.

ITS

Refer to Internationalization Tag Set (ITS).

Java

Java is a programming language that is also used to implement LanguageTool. Refer to http://www.java.com.

Javadoc

Javadoc is software from Oracle that uses comments in the source code to create API documentation in HTML format. Refer to http://www.oracle.com/technetwork/java/javase/documentation/index-jsp-135444.html.

language code

A language code is a code that represents a language, and possibly, a variant of a language. For example, the language code for German is de. Different standards for language codes exist. LanguageTool uses IETF language tags.

language maintainer

A language maintainer is a committer who maintains the rules in LanguageTool for one or more languages and who translates the LanguageTool GUI.

lemma

A lemma is the 'base form' of a word or of a group of words that is one lexical unit (lexeme). English examples: help, laser printer, put up with.

LGPL

LGPL (GNU Lesser General Public License) is a licence to use software. Refer to http://www.gnu.org/licenses/lgpl.html.

Lucene

Search engine library used by internally LanguageTool for some of the Wikipedia-related features. See http://lucene.apache.org.

markup language

A markup language is a computer language that uses special text to specify the structure or the style of the content of a document. Refer to element; XML.

Maven

Maven is a 'build automation tool' for software developers. Maven compiles and packages software, and manages dependencies. Refer to http://maven.apache.org.

Morfologik

the Morfologik stemming library is used by LT for dictionary lookups. Morfologik allows compressing large text dictionaries into small binary files with fast word lookup. https://github.com/morfologik/morfologik-stemming.

noun chunk (noun phrase)

A noun chunk is one or more words that acts as a noun. Refer to http://wiki.languagetool.org/using-chunks. LanguageTool identifies noun chunks using a chunker. Refer to http://www.inf.ed.ac.uk/teaching/courses/icl/lectures/chunk.2up.pdf

noun phrase

Refer to noun chunk.

part-of-speech tag

Refer to POS tag.

part-of-speech tagger

Refer to POS tagger.

POS

POS is an abbreviation for part of speech.

POS tag (part-of-speech tag)

A POS tag is a tag that identifies the possible parts of speech that a word has. Usually, in LanguageTool, a part-of-speech tag also includes other information, such as whether a noun is singular or plural.

POS tagger

A POS tagger is software that annotates (tags) words with part-of-speech information. A POS tagger gives each word one or more POS tags. Refer to http://wiki.languagetool.org/developing-a-tagger-dictionary.

property file

[?]

reading

http://wiki.languagetool.org/developing-a-disambiguator. Many English words can be, for example, a verb or a noun, depending on their context, so they have to readings. Example: "walk" can be a verb ("I walk home") or a noun ("I took a walk").

regular expression

A regular expression is text that specifies a search pattern. Refer to http://www.regular-expressions.info.

rulegroup

A rulegroup is an element in the grammar.xml file. If more than one rule is necessary to find an error, you can put all the applicable rules into a rulegroup. Refer to http://wiki.languagetool.org/development-overview#toc9.

shallow parser

The term shallow parser is a synonym for the term chunker.

skip (verb)

To skip means optionally to ignore one or more tokens in a sequence of tokens. For example, possibly you want to ignore a punctuation mark if it exists in a sequence of text. Refer to http://wiki.languagetool.org/development-overview#toc13.

SRX

SRX (Segmentation Rules eXchange) is a method of specifying the segmentation rules that software uses to split text into segments. (Typically, a segment is equivalent to a sentence.) Refer to http://www.gala-global.org/oscarStandards/srx/srx20.html.

tagger

Refer to POS tagger.

testrules.sh/testrules.bat

testrules is software that you can use to make sure that the rules that you write in LanguageTool are correct. Refer to http://wiki.languagetool.org/development-overview.

Tika

Tika™ is software that detects and extracts metadata (data about data) and structured text content from documents. Tika is supplied by Apache. Refer to http://tika.apache.org.

token

A token is an instance of a sequence of characters in some particular document that are grouped together as a useful semantic unit for processing (http://nlp.stanford.edu/IR-book/html/htmledition/tokenization-1.html). Typically, in LanguageTool, a token is equivalent to a word. However, punctuation marks are also tokens.

Transifex

Transifex is an online translation-management system. The translations for LanguageTool are on https://www.transifex.com/projects/p/languagetool/.

unification

Unification is the matching of sequences of tokens that have some specified features. Refer to http://wiki.languagetool.org/using-unification.

unify

Refer to unification.

verb chunk

[In LT, what is a verb chunk? http://en.wikipedia.org/wiki/Verb_phrase]

verb phrase

Refer to verb chunk.

XML

XML is a markup language. Refer to http://xml.coverpages.org/xml.html. In LanguageTool, the language rules are specified using XML. Refer to disambiguation.xml; grammar.xml;