... ambiguity Statistical machine translation (SMT) has been proposed as a means of context-sensitive text normalisation, by treating the ill-formed text as the source language, and the standard form as the ... 145–148, Sapporo, Japan Joseph Kaufmann and Jugal Kalita 2010 Syntactic normalization of Twitter messages In International Conference on Natural Language Processing, Kharagpur, India Dan Klein and ... develop a method which does not require annotated training data, but is able to leverage context for lexical normalisation Our approach first generates a list of candidate canonical lexical forms, based...