论文信息 - Bekli: A Simple Approach to Twitter Text Normalization

Bekli: A Simple Approach to Twitter Text Normalization

Every day, Twitter users generate vast quantities of potentially useful information in the form of written language. Due to Twitter’s frequently informal tone, text normalization can be a crucial element for exploiting that information. This paper outlines our approach to text normalization used in the WNUT shared task. We show that a very simple solution, powered by a modestly sized, partiallycurated wordlist—combined with a modest reranking scheme—can deliver respectable results.

Russell Beckley

[1] Yi Yang,et al. A Log-Linear Model for Unsupervised Text Normalization , 2013, EMNLP.

[2] Shankar Kumar,et al. Normalization of non-standard words , 2001, Comput. Speech Lang..

[3] Timothy Baldwin,et al. Shared Tasks of the 2015 Workshop on Noisy User-generated Text: Twitter Lexical Normalization and Named Entity Recognition , 2015, NUT@IJCNLP.

[4] Eduard H. Hovy,et al. Weakly Supervised User Profile Extraction from Twitter , 2014, ACL.

[5] Junlan Feng,et al. Robust Sentiment Detection on Twitter from Biased and Noisy Data , 2010, COLING.

[6] Timothy Baldwin,et al. Lexical Normalisation of Short Text Messages: Makn Sens a #twitter , 2011, ACL.

[7] Animesh Mukherjee,et al. Investigation and modeling of the structure of texting language , 2007, International Journal of Document Analysis and Recognition (IJDAR).

[8] Preslav Nakov,et al. SemEval-2015 Task 10: Sentiment Analysis in Twitter , 2015, *SEMEVAL.

[9] Jennifer Foster,et al. Sentiment Analysis of Political Tweets: Towards an Accurate Classifier , 2013 .

[10] François Yvon,et al. Normalizing SMS: are Two Metaphors Better than One ? , 2008, COLING.

[11] G. Eysenbach,et al. Pandemics in the Age of Twitter: Content Analysis of Tweets during the 2009 H1N1 Outbreak , 2010, PloS one.

[12] Yang Liu,et al. Improving Text Normalization via Unsupervised Model and Discriminative Reranking , 2014, ACL.