FreeLing is an open-source library providing a wide range of language analysis utilities for
several different languages. It is intended to provide NLP application developers with any text
processing and language annotation tools they may need in order to simplify their development
task.
Moreover, FreeLing is customizable and extensible. Developers can use the default linguistic
resources (dictionaries, lexicons, grammars, etc.), or extend them, adapt to particular domains, or
even develop new resources for specific languages.
Being open-source has enabled FreeLing to grow far beyond its original capabilities,
especially with regard to linguistic data: contributions from its community of users, for instance,
include morphological dictionaries and PoS tagger training data for Galician, Italian, Portuguese,
Asturian, and Welsh.
In this paper we present the basic architecture and the main services in FreeLing, and we
outline how developers might use it to build competitive NLP systems and indicate how it might
be extended to support the development of Example-Based Machine Translation systems.
[1]
Aingeru Mayor,et al.
TXALA un analizador libre de dependencias para el castellano
,
2005,
Proces. del Leng. Natural.
[2]
Eneko Agirre,et al.
Personalizing PageRank for Word Sense Disambiguation
,
2009,
EACL.
[3]
Hwee Tou Ng,et al.
A Machine Learning Approach to Coreference Resolution of Noun Phrases
,
2001,
CL.
[4]
Thorsten Brants,et al.
TnT – A Statistical Part-of-Speech Tagger
,
2000,
ANLP.
[5]
Xavier Carreras,et al.
Named Entity Extraction using AdaBoost
,
2002,
CoNLL.
[6]
Lluís Padró,et al.
A Hybrid Environment for Syntax-Semantic Tagging
,
1998,
ArXiv.