Named Entity Recognition in Tweets: An Experimental Study

Proceedings of the 2011 Conference on Empirical Methods in Natural Language Processing, pages 1524–1534, Edinburgh, Scotland, UK, July 27–31, 2011. c⃝2011 Association for Computational Linguistics Named Entity Recognition in Tweets: An Experimental Study Alan Ritter, Sam Clark, Mausam and Oren Etzioni Computer Science and Engineering University of Washington Seattle, WA 98125, USA {aritter,ssclark,mausam,etzioni}@cs.washington.edu Abstract People tweet more than 100 Million times daily, yielding a noisy, informal, but some- times informative corpus of 140-character messages that mirrors the zeitgeist in an un- precedented manner. The performance of standard NLP tools is severely degraded on tweets. This paper addresses this issue by re-building the NLP pipeline beginning with part-of-speech tagging, through chunking, to named-entity recognition. Our novel T-NER system doubles F1 score compared with the Stanford NER system. T-NER leverages the redundancy inherent in tweets to achieve this performance, using LabeledLDA to exploit Freebase dictionaries as a source of distant supervision. LabeledLDA outperforms co- training, increasing F1 by 25% over ten com- mon entity types. Our NLP tools are available at: http:// github.com/aritter/twitter_nlp 1 Introduction Status Messages posted on Social Media websites such as Facebook and Twitter present a new and challenging style of text for language technology due to their noisy and informal nature. Like SMS (Kobus et al., 2008), tweets are particularly terse and difficult (See Table 1). Yet tweets provide a unique compilation of information that is more up- to-date and inclusive than news articles, due to the low-barrier to tweeting, and the proliferation of mo- bile devices.1 The corpus of tweets already exceeds 1See the “trending topics” displayed on twitter.com the size of the Library of Congress (Hachman, 2011) and is growing far more rapidly. Due to the vol- ume of tweets, it is natural to consider named-entity recognition, information extraction, and text mining over tweets. Not surprisingly, the performance of “off the shelf” NLP tools, which were trained on news corpora, is weak on tweet corpora. In response, we report on a re-trained “NLP pipeline” that leverages previously-tagged out-of- domain text, 2 tagged tweets, and unlabeled tweets to achieve more effective part-of-speech tagging, chunking, and named-entity recognition. 1 The Hobbit has FINALLY started filming! I cannot wait! 2 Yess! Yess! Its official Nintendo announced today that they Will release the Nintendo 3DS in north America march 27 for $250 3 Government confirms blast n nuclear plants n japan...don’t knw wht s gona happen nw... Table 1: Examples of noisy text in tweets. We find that classifying named entities in tweets is a difficult task for two reasons. First, tweets contain a plethora of distinctive named entity types (Compa- nies, Products, Bands, Movies, and more). Almost all these types (except for People and Locations) are relatively infrequent, so even a large sample of man- ually annotated tweets will contain few training ex- amples. Secondly, due to Twitter’s 140 character limit, tweets often lack sufficient context to deter- mine an entity’s type without the aid of background 2Although tweets can be written on any subject, following convention we use …

Content truncated to 500 words