brWaC: A WaCky Corpus for Brazilian Portuguese

摘要

Initiatives for constructing very large corpora have increased in recent years, especially using the Web as corpus since large corpora are crucial for many Natural Language Processing tasks. The WaCky (Web-As-Corpus Kool Yinitiative) methodology has been used to build very large corpora (over a billion words each) for languages like English, Italian and German among others. In this paper we present the ongoing work on building brWaC, a massive Brazilian Portuguese corpus crawled from .br domains. At the moment, the crawling process and the PoS tagging are finished, resulting in a tokenized and lemmatized corpus of 3 billion words. Next step is parsing the whole corpus.