FS-NER: a lightweight filter-stream approach to named entity recognition on twitter data

Microblog platforms such as Twitter are being increasingly adopted by Web users, yielding an important source of data for web search and mining applications. Tasks such as Named Entity Recognition are at the core of many of these applications, but the effectiveness of existing tools is seriously compromised when applied to Twitter data, since messages are terse, poorly worded and posted in many different languages. Also, Twitter follows a streaming paradigm, imposing that entities must be recognized in real-time. In view of these challenges and the inappropriateness of existing tools, we propose a novel approach for Named Entity Recognition on Twitter data called FS-NER (Filter-Stream Named Entity Recognition). FS-NER is characterized by the use of filters that process unlabeled Twitter messages, being much more practical than existing supervised CRF-based approaches. Such filters can be combined either in sequence or in parallel in a flexible way. Moreover, because these filters are not language dependent, FS-NER can be applied to different languages without requiring a laborious adaptation. Through a systematic evaluation using three Twitter collections and considering seven types of entity, we show that FS-NER performs 3% better than a CRF-based baseline, besides being orders of magnitude faster and much more practical.