?url_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Adc&rft.title=Targetable+Named+Entity+Recognition+in+Social+Media&rft.creator=Ashwini%2C+Sandeep&rft.creator=Choi%2C+Jinho+D.&rft.subject=Language&rft.subject=Comparative+Linguistics&rft.description=We+present+a+novel+approach+for+recognizing+what+we+call+targetable+named+entities%3B+that+is%2C+named+entities+in+a+targeted+set+(e.g%2C+movies%2C+books%2C+TV+shows).+Unlike+many+other+NER+systems+that+need+to+retrain+their+statistical+models+as+new+entities+arrive%2C+our+approach+does+not+require+such+retraining%2C+which+makes+it+more+adaptable+for+types+of+entities+that+are+frequently+updated.+For+this+preliminary+study%2C+we+focus+on+one+entity+type%2C+movie+title%2C+using+data+collected+from+Twitter.+Our+system+is+tested+on+two+evaluation+sets%2C+one+including+only+entities+corresponding+to+movies+in+our+training+set%2C+and+the+other+excluding+any+of+those+entities.+Our+final+model+shows+F1-scores+of+76.19%25+and+78.70%25+on+these+evaluation+sets%2C+which+gives+strong+evidence+that+our+approach+is+completely+unbiased+to+any+particular+set+of+entities+found+during+training.&rft.date=2014-07-31&rft.type=Preprint&rft.type=NonPeerReviewed&rft.format=application%2Fpdf&rft.identifier=http%3A%2F%2Fcogprints.org%2F9764%2F1%2Fcogprints2014a.pdf&rft.identifier=++Ashwini%2C+Sandeep+and+Choi%2C+Jinho+D.++(2014)+Targetable+Named+Entity+Recognition+in+Social+Media.++%5BPreprint%5D++++(Unpublished)++&rft.relation=http%3A%2F%2Fcogprints.org%2F9764%2F