?url_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Adc&rft.title=Towards+Incremental+Parsing+of+Natural+Language+using+Recursive+Neural+Networks&rft.creator=Costa%2C+Fabrizio&rft.creator=Frasconi%2C+Paolo&rft.creator=Lombardo%2C+Vincenzo&rft.creator=Soda%2C+Giovanni&rft.subject=Language&rft.subject=Machine+Learning&rft.subject=Neural+Nets&rft.description=In+this+paper+we+develop+novel+algorithmic+ideas+for+building+a+natural+language%0Aparser+grounded+upon+the+hypothesis+of+incrementality.+Although+widely+accepted%0Aand+experimentally+supported+under+a+cognitive+perspective+as+a+model+of+the+human%0Aparser%2C+the+incrementality+assumption+has+never+been+exploited+for+building+automatic%0Aparsers+of+unconstrained+real+texts.+The+essentials+of+the+hypothesis+are+that+words+are%0Aprocessed+in+a+left-to-right+fashion%2C+and+the+syntactic+structure+is+kept+totally+connected%0Aat+each+step.%0AOur+proposal+relies+on+a+machine+learning+technique+for+predicting+the+correctness+of%0Apartial+syntactic+structures+that+are+built+during+the+parsing+process.+A+recursive+neural%0Anetwork+architecture+is+employed+for+computing+predictions+after+a+training+phase+on%0Aexamples+drawn+from+a+corpus+of+parsed+sentences%2C+the+Penn+Treebank.+Our+results%0Aindicate+the+viability+of+the+approach+andlay+out+the+premises+for+a+novel+generation+of%0Aalgorithms+for+natural+language+processing+which+more+closely+model+human+parsing.%0AThese+algorithms+may+prove+very+useful+in+the+development+of+e%C3%86cient+parsers.&rft.publisher=Kluwer+academic+publishers&rft.date=2002&rft.type=Journal+(Paginated)&rft.type=PeerReviewed&rft.format=application%2Fpdf&rft.identifier=http%3A%2F%2Fcogprints.org%2F2089%2F3%2Fai-2001-Towards.pdf&rft.identifier=++Costa%2C+Fabrizio+and+Frasconi%2C+Paolo+and+Lombardo%2C+Vincenzo+and+Soda%2C+Giovanni++(2002)+Towards+Incremental+Parsing+of+Natural+Language+using+Recursive+Neural+Networks.++%5BJournal+(Paginated)%5D++++(In+Press)++&rft.relation=http%3A%2F%2Fcogprints.org%2F2089%2F