IntroductionWhy doesn't text search (IR)work?What you search for in real estate advertisements:Town/suburb. You might think easy, but:Realestateagents:ColdwellBanker,MosmanPhrases:Only45minutesfromParramattaMultiplepropertyadshavedifferent suburbs inoneadMoney: want a range not a textual matchMultipleamounts:was$155K,now$145K.Variations: offers in the high 700s [but not rents for $270]
Introduction What you search for in real estate advertisements: • Town/suburb. You might think easy, but: • Real estate agents: Coldwell Banker, Mosman • Phrases: Only 45 minutes from Parramatta • Multiple property ads have different suburbs in one ad • Money: want a range not a textual match • Multiple amounts: was $155K, now $145K • Variations: offers in the high 700s [but not rents for $270] l Why doesn’t text search (IR) work?
IntroductionWhy doesn't text search (IR)work?What you search for in real estate advertisements:Town/suburb. You might think easy, but:Realestateagents:ColdwellBanker,MosmanPhrases:Only45minutesfromParramattaMultiplepropertyadshavedifferent suburbs inoneadMoney: want a range not a textual match.Multipleamounts:was$155K,now$145K. Variations: offers in the high 700s [but not rents for $270]Bedrooms:similarissues:br, bdr, beds, B/R
Introduction What you search for in real estate advertisements: • Town/suburb. You might think easy, but: • Real estate agents: Coldwell Banker, Mosman • Phrases: Only 45 minutes from Parramatta • Multiple property ads have different suburbs in one ad • Money: want a range not a textual match • Multiple amounts: was $155K, now $145K • Variations: offers in the high 700s [but not rents for $270] • Bedrooms: similar issues: br, bdr, beds, B/R l Why doesn’t text search (IR) work?
Introduction（NER+RE）InformationExtraction交道大学
l Information Extraction （NER+ RE） Introduction
IntroductionInformationExtraction（NER+RE)NamedEntity:thewordorphrasethatrepresentsaspecificreal-worldobjectNamed entityrecognition(NER):thetaskofidentifyingallthementions(occurrences)ofaparticularNEtypeinthegivendocumentsDetection:Mr.Smitheatsbitterballen[Mr.Smith] :ENTITYClassification:Mr.Smitheatsbitterballen[Mr.Smith]:PERSON支通大学
● Named Entity： the word or phrase that represents a specific real-world object. ● Named entity recognition (NER): the task of identifying all the mentions (occurrences) of a particular NE type in the given documents. Detection: Mr. Smith eats bitterballen [Mr. Smith] : ENTITY Classification: Mr. Smith eats bitterballen [Mr. Smith] : PERSON l Information Extraction （NER+ RE） Introduction
IntroductionInformationExtraction（NER+RE)Named Entity:the word orphrasethat represents a specific real-world object.Named entity recognition (NER):thetaskofidentifying all the mentions(occurrences)ofaparticularNEtypeinthegivendocumentsDetection:Mr.Smith eats bitterballen[Mr.Smith] :ENTITYClassification:Mr.Smitheatsbitterballen[Mr.Smith] :PERSONRelation:usually denotes a well-defined (having a specific meaning)relationshipbetweentwoormoreNEsRelationship Extraction(RE):identifymentionsoftherelations of interestin eachsentenceofthegivendocuments交道大学
● Named Entity： the word or phrase that represents a specific real-world object. ● Named entity recognition (NER): the task of identifying all the mentions (occurrences) of a particular NE type in the given documents. Detection: Mr. Smith eats bitterballen [Mr. Smith] : ENTITY Classification: Mr. Smith eats bitterballen [Mr. Smith] : PERSON ● Relation: usually denotes a well-defined (having a specific meaning) relationship between two or more NEs. ● Relationship Extraction (RE): identify mentions of the relations of interest in each sentence of the given documents. l Information Extraction （NER+ RE） Introduction