?url_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Adc&rft.title=Word+Sense+Disambiguation+by+Web+Mining+for+Word+Co-occurrence+Probabilities&rft.creator=Turney%2C+Peter+D.&rft.subject=Language&rft.subject=Computational+Linguistics&rft.subject=Semantics&rft.subject=Machine+Learning&rft.description=This+paper+describes+the+National+Research+Council+(NRC)%0AWord+Sense+Disambiguation+(WSD)+system%2C+as+applied+to+the%0AEnglish+Lexical+Sample+(ELS)+task+in+Senseval-3.+The+NRC+system+%0Aapproaches+WSD+as+a+classical+supervised+machine+learning+problem%2C%0Ausing+familiar+tools+such+as+the+Weka+machine+learning+software+%0Aand+Brill's+rule-based+part-of-speech+tagger.+Head+words+are%0Arepresented+as+feature+vectors+with+several+hundred+features.%0AApproximately+half+of+the+features+are+syntactic+and+the+other%0Ahalf+are+semantic.+The+main+novelty+in+the+system+is+the+method+for%0Agenerating+the+semantic+features%2C+based+on+word+co-occurrence+%0Aprobabilities.+The+probabilities+are+estimated+using+%0Athe+Waterloo+MultiText+System+with+a+corpus+of+about+one+terabyte+of+%0Aunlabeled+text%2C+collected+by+a+web+crawler.&rft.date=2004&rft.type=Conference+Paper&rft.type=PeerReviewed&rft.format=application%2Fpdf&rft.identifier=http%3A%2F%2Fcogprints.org%2F3732%2F1%2FNRC-47167.pdf&rft.identifier=++Turney%2C+Peter+D.++(2004)+Word+Sense+Disambiguation+by+Web+Mining+for+Word+Co-occurrence+Probabilities.++%5BConference+Paper%5D+++++&rft.relation=http%3A%2F%2Fcogprints.org%2F3732%2F