IntroductionRelationshipExtractionRelationship Extraction:identifymentions ofthe relations ofinterest in eachsentenceofthegivendocuments(IBMortheExample:"InternationalBusinessMachinesCorporationcompany)was incorporated intheStateof NewYorkonJune16,1911,astheComputing-Tabulating-RecordingCo.(C-T-R)..."交通大学
● Relationship Extraction: identify mentions of the relations of interest in each sentence of the given documents. Example: “International Business Machines Corporation (IBM or the company) was incorporated in the State of New York on June 16, 1911, as the Computing-Tabulating-RecordingCo. (C-T-R).” l Relationship Extraction Introduction
IntroductionRelationshipExtractionRelationship Extraction:identifymentions ofthe relations of interest in eachsentenceofthegivendocuments.(IBMExample:"InternationalBusinessMachinesCorporationorthecompany)was incorporated intheState of NewYorkonJune16,1911,astheComputing-Tabulating-RecordingCo.(C-T-R)..."ExtractedComplexRelation:Company-FoundingIBMCompanyLocationNew YorkDateJune 16, 1911Original-NameComputing-Tabulating-Recording CoThesimplertaskofextractingrelationtriples (focuson)Founding-year(IBM, 1911)Founding-location(IBM,NewYork)
● Relationship Extraction: identify mentions of the relations of interest in each sentence of the given documents. Example: “International Business Machines Corporation (IBM or the company) was incorporated in the State of New York on June 16, 1911, as the Computing-Tabulating-RecordingCo. (C-T-R).” Extracted Complex Relation: Company-Founding Company IBM Location New York Date June 16, 1911 Original-Name Computing-Tabulating-Recording Co. The simpler task of extracting relation triples (focus on) Founding-year(IBM, 1911) Founding-location(IBM, New York) l Relationship Extraction Introduction
IntroductionRelationshipExtractionRelationship Extraction:identifymentions ofthe relations of interest in eachsentence ofthegivendocuments.(IBMExample:"InternationalBusinessMachinesCorporationorthecompany)was incorporated intheState of NewYorkonJune16,1911,astheComputing-Tabulating-RecordingCo.(C-T-R)..."ExtractedComplexRelation:Company-FoundingIBMCompanyLocationNewYorkDateJune 16, 1911Original-NameComputing-Tabulating-Recording CoThesimplertaskofextractingrelationtriples (focuson)Founding-year(IBM, 1911)Founding-location(IBM,New York) Note: it is possible to treat relation tagging as a classification problem, classifyingeachpairas a relationtype or NONE
● Relationship Extraction: identify mentions of the relations of interest in each sentence of the given documents. Example: “International Business Machines Corporation (IBM or the company) was incorporated in the State of New York on June 16, 1911, as the Computing-Tabulating-RecordingCo. (C-T-R).” Extracted Complex Relation: Company-Founding Company IBM Location New York Date June 16, 1911 Original-Name Computing-Tabulating-Recording Co. The simpler task of extracting relation triples (focus on) Founding-year(IBM, 1911) Founding-location(IBM, New York) ● Note: it is possible to treat relation tagging as a classification problem, classifying each pair as a relation type or NONE. l Relationship Extraction Introduction
IntroductionRelationshipExtractionandCBF-CinteractCBF-A“"WeshowthatwitheachothertoformaICBF-A-CBF-Ccomplexand thatCBF-Bdoesnot interact with CBF-A or CBF-C individuallybut that itassociateswith theCBF-A-CBF-Ccomplex.interactCBF-ACBF-CcomplexCBF-B*CBF-A-CBF-C complexassociates
“We show that CBF-A and CBF-C interact with each other to form a CBF-A-CBF-C complex and that CBF-B does not interact with CBF-A or CBF-C individually but that it associates with the CBF-A-CBF-C complex.“ CBF-A CBF-C CBF-B CBF-A-CBF-C complex interact complex associates l Relationship Extraction Introduction
IntroductionMaingoals ofrelationextractionFill a predefinedtemplatefromrawtextExtractwho did whatto whom and when?EventextractionOrganize information so that is useful to peoplePut information ina form that allows furtherinferences by computersBigdataCreate new structured knowledge bases, useful for any appAugmentcurrentknowledgebasesSuchas:WordNet,FreeBase,DBPedia,BiomodelsSupport question answering夷通大学
• Fill a predefined “template” from raw text • Extract who did what to whom and when? Event extraction • Organize information so that is useful to people • Put information in a form that allows further inferences by computers Big data • Create new structured knowledge bases, useful for any app • Augment current knowledge bases Such as: WordNet, FreeBase, DBPedia, Biomodels • Support question answering l Main goals of relation extraction Introduction