LORELEI Arabic Representative Language Pack

Item Name: LORELEI Arabic Representative Language Pack
Author(s): Jennifer Tracey, Stephanie Strassel, David Graff, Jonathan Wright, Song Chen, Neville Ryant, Seth Kulick, Kira Griffitt, Dana Delgado, Michael Arrigo
LDC Catalog No.: LDC2026T08
ISLRN: 919-721-145-167-6
DOI: https://doi.org/10.35111/a4jq-jq94
Release Date: August 17, 2026
Member Year(s): 2026
DCMI Type(s): Software, Text
Data Source(s): discussion forum, newswire, web collection, weblogs
Project(s): LORELEI
Application(s): cross-language transfer, entity extraction, information extraction, machine translation
Language(s): Arabic, English
Language ID(s): ara, eng
License(s): LDC User Agreement for Non-Members
Online Documentation: LDC2026T08 Documents
Licensing Instructions: Subscription & Standard Members, and Non-Members
Citation: Tracey, Jennifer, et al. LORELEI Arabic Representative Language Pack LDC2026T08. Web Download. Philadelphia: Linguistic Data Consortium, 2026.
Related Works: View

Introduction

LORELEI Arabic Representative Language Pack consists of Arabic monolingual text, Arabic-English parallel text, annotations, supplemental resources and related software tools developed by the Linguistic Data Consortium for the DARPA LORELEI program.

The LORELEI (Low Resource Languages for Emergent Incidents) program was concerned with building human language technology for low resource languages in the context of emergent situations like natural disasters or disease outbreaks. Linguistic resources for LORELEI include Representative Language Packs and Incident Language Packs for over two dozen low resource languages, comprising data, annotations, basic natural language processing tools, lexicons and grammatical resources. Representative languages were selected to provide broad typological coverage, while incident languages were selected to evaluate system performance on a language whose identity was disclosed at the start of the evaluation.

Data

Arabic is spoken throughout the Middle East and North Africa and is an official language in more than 20 countries. Data was collected in the following genres: discussion forum, news, weblogs and social network. Both monlingual text collection and parallel text creation involved a combination of manual and automatic methods.

Data volumes are as follows:

  • 2.4 million words of Arabic monolingual text, over 930,000 words of which were translated into English
  • 225,000 Arabic words translated from English data

Nearly 88,000 words were annotated for simple named entities, and approximately 29,000 words were annotated for full entities (including nominals and pronouns). In addition, over 10,000 words were annotated for noun phrase chunking, 18,000 words were labeled with semantic annotation and situation frame annotation was applied to more than 24,000 words.

Lexical resources and software tools are also included in this release. The tools recreate original source data from the processed XML material, condition text data users download from Twitter/X, apply sentence segmentation to raw text, and support named entity tagging.

Monolingual and parallel text are presented in XML with associated DTDs. Annotation data is presented as XML or tab-delimited files. All text is UTF-8 encoded.

The knowledge base for entity linking annotation for this corpus and all LORELEI Representative Language and Incident Language Packs is available separately as LORELEI Entity Detection and Linking Knowledge Base (LDC2020T10).

Sponsorship

This material is based upon work supported by the Defense Advanced Research Projects Agency (DARPA) under Contract No. HR0011-15-C-0123. Any opinions, findings and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of DARPA.

Samples

Please view these samples:

Updates

No updates at this time.

Available Media

View Fees





Login for the applicable fee