... results on the task of au-tomatic dialect identification, using the col-lected labels for training and evaluation.1 Introduction The Arabic language is characterized by an interest-ing linguistic ... Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics:shortpapers, pages 37–41,Portland, Oregon, June 19-24, 2011.c2011 Association for Computational Linguistics The ... Thisresulted in crawling about 150K URL’s, 86.1K of which included reader commentary (Table 1). The data consists of 1.4M comments, corresponding to52.1M words.We also extract the following information...