To Normalize, or Not to Normalize: The Impact of Normalization on Part-of-Speech Tagging
This addresses the problem of improving POS tagging for noisy text like tweets, but it is incremental as it compares existing methods without major breakthroughs.
The study investigated whether normalization improves Part-of-Speech tagging accuracy on noisy tweet data, finding that normalization helps but does not consistently outperform using word embeddings without normalization, which achieved competitive results with a state-of-the-art Twitter tagger.
Does normalization help Part-of-Speech (POS) tagging accuracy on noisy, non-canonical data? To the best of our knowledge, little is known on the actual impact of normalization in a real-world scenario, where gold error detection is not available. We investigate the effect of automatic normalization on POS tagging of tweets. We also compare normalization to strategies that leverage large amounts of unlabeled data kept in its raw form. Our results show that normalization helps, but does not add consistently beyond just word embedding layer initialization. The latter approach yields a tagging model that is competitive with a Twitter state-of-the-art tagger.