邹磊 - 图数据库技术在知识图谱数据管理中的应用

羊舌凝绿

2018/05/13 发布于 技术 分类

近年来,随着“人工智能”概念再度活跃,除了“深度学习”这个炙手可热的名词以外,“知识图谱”也是研究者、工业界和投资人心目中的又一颗“银弹”。简单地说,“知识图谱”是一种数据模型,是以图形(Graph)的方式来展现“实体”、实体“属性”,以及实体之间的“关系”。图1所示的例子中有4个实体,分别是“达·芬奇”“意大利”“蒙拉丽莎”和“米开朗基罗”。这个图明确地展示了“达·芬奇”的各个属性和属性值(例如名字、生日和逝世时间等),以及这4个实体之间的关系(例如蒙拉丽莎是达·芬奇的画作,达·芬奇出生在意大利等)。

文字内容
1. Graph-based RDF Data Managment Lei Zou Peking University Institute of Computer Science and Technology
2. RDF and Semantic Web RDF is a language for the conceptual modeling of information about web resources A building block of semantic web Facilitates exchange of information Search engines can retrieve more relevant information Facilitates data integration (mashes) Machine understandable Understand the information on the web and the interrelationships among them
3. What’s Sematic Web: A Simple Example (RDFa) The traditional Web (HTML) only considers the display of the content. How is the page displayed, such as which font and the format of the pictures ? Lei Zou
Email: zoulei@pku.edu.cn

Publications:

Lei Zou, Jinhui Mo, Lei Chen, M. Tamer Ozsu, Dongyan Zhao, gStore: Answering SPARQL Queries Via Subgraph Matching, VLDB, 2011
4. What’s Sematic Web: A Simple Example (RDFa) Sematic Web considers the sematics of the content. What does the content in the page mean? e.g., What are the mean of “zoulei@pku.edu.cn” and “VLDB” ?
http://xmlns.com/foaf/0.1/name> Lei Zou
zoulei@pku.edu.cn

Publications:

Lei Zou , Jinghui Mo , Lei Chen , M. Tamer Özsu, Dongyan Zhao, gStore: Answering SPARQL Queries Via Subgraph Matching , VLDB 2011
5. What’s Sematic Web: Google Snippet
6. What’s Sematic Web: Google Snippet
7. What’s Sematic Web: Google Snippet
8. What’s Sematic Web: Google Snippet
9. What’s Sematic Web: Facebook Social Graph
10. What’s Sematic Web: From Two Perspectives Expressiveness More Semantic; More Powerful Reasoning RDF, RDFS,OWL,OWL Full, OWL 2 …… Open Linked Data, Web-scale Triple Store, Semantic Wiki …… How to get more data ? How to manage the Web-scale Semantic Data ? Scalability
11. What’s Sematic Web: From Two Perspectives More Semantic; More Powerful Reasoning RDF, RDFS,OWL,OWL Full, OWL 2 …… Expressiveness More Interesting Applications Apple Siri, Google Knowledge Graph; IBM Watson; …… More Areas Broadcasting: BBC Publishing: Thomson Reuters Life: Eli Lilly and Company …… Open Linked Data, Web-scale Triple Store, Semantic Wiki …… How to get more data ? How to manage the Web-scale Semantic Data ? Scalability
12. Some Interesting Products IBM Watson
13. Some Interesting Products EVI— acquired by Amazon on October 2012. William Tunstall-Pedoe: True Knowledge: Open-Domain Question Answering Using Structured Knowledge and Inference. AI Magazine 31(3): 80-92 (2010)
14. Some Interesting Products Google Knowledge Graph
15. RDF Uses Yago and DBPedia extract facts from Wikipedia & represent as RDF → structural queries Communities build RDF data E.g., biologists: Bio2RDF and Uniprot RDF Web data integration Linked Data Cloud ...
16. RDF Data Volumes . . . . . . are growing – and fast Linked data cloud currently consists of 325 datasets with >25B triples Size almost doubling every year
17. RDF Data Volumes . . . . . . are growing – and fast Linked data cloud currently consists of 325 datasets with >25B triples Size almost doubling every year MySpace Wrapper Musicbrainz Surge Radio LIBRIS AudioScrobbler QDOS Doapspace SemWebCentral Wikicompany Flickr exporter Semantic Web.org RDF ohloh SW Conference Corpus Resex IRIT Toulouse Eurécom BBC Playcount Data Jamendo BBC Later + TOTP BBC John Peel Crunch Base FOAF profiles SIOC Sites Revyu OpenGuides ACM DBLP BBC Pub Guide Geonames Eurostat Project Guten- berg flickr wrappr Virtuoso Sponger Pisa RKB Explorer Programm es Open Calais GovTrack riese Magnatune World Factbook DBpedia Linked MDB RDF Book Mashup ECS Southampton IEEE US Census Data W3C WordNet UMBEL Open Cyc Yago Homolo Gene Pub Chem Daily Med lingvoj Freebase LinkedCT GEO Species Drug Bank KEGG DBLP Berlin GeneID DBLP Hannover CiteSeer UniRef Reactome UniParc PROSITE UniProt RAE 2001 Budapest BME eprints Newcastle IBM LAAS- CNRS Taxonomy Symbol Diseasome UniSTS CAS OMIM ChEBI Gene Ontology HGNC MGI PubMed Inter Pro Pfam PDB ProDom As of March 2009 March ’09: 89 datasets Linking Open Data cloud diagram, by Richard Cyganiak and Anja Jentzsch. http://lod-cloud.net/
18. RDF Data Volumes . . . . . . are growing – and fast Linked data cloud currently consists of 325 datasets with >25B triples Size almost doubling every year Sussex St. MySpace Audioscrobbler Reading Lists Andrews Resource Lists NDL subjects t4gm biz. data. gov.uk Energy (En- AKTing) Population (EnAKTing) NHS (EnAKTing) CO2 (EnAKTing) Mortality (En- AKTing) legislation .gov.uk UK Postcodes Ordnance Survey ESD standards reference data.gov .uk Linked Data for Intervals The London Gazette TWC LOGD GTAA DB Tropes Magnatune John Peel (DB Tune) research data.gov .uk FanHubz Surge Radio EUTC Produc- tions Last.fm Artists (DBTune) Moseley Folk (DBTune) Music Brainz (Data Incubator) Discogs (Data Incubator) Music Brainz (DBTune) Last.FM (rdfize) (DBTune) Music Brainz (zitgist) Plymouth Reading Lists Manchester Reading Lists The Open Library NTU Resource Lists The Open Library (Talis) Jamendo Poképédia Linked LCCN classical (DB Pokedex LIBRIS VIAF LCSH RAMEAU SH lobid Organisations lobid Resources PSH Gem. Norm- datei RDF Book P20 Mashup UB Mannheim semantic web.org ECS Southampton EPrints ECS Southampton Pisa Ulm RISKS DEPLOY RESEX Wiki Eurécom Tune) PBAC ECS BBC education data.gov .uk OpenEI Program mes BBC BBC Music Chronicling America EventMedia Linked MDB NSZL Catalog Openly Local Wildlife Finder Rechtspraak. nl Telegraphis New York Times flickr URI wrappr Open statistics Burner Calais data.gov .uk LOIUS Taxon Concept NASA Geo Names World Factbook (FUB) Freebase transport data.gov .uk (Data Incubator) Eurostat Fishes of Texas Geo Species Uberblic DBpedia dbpedia (FUB) Geo lite Eurostat Linked Data (es) UMBEL lingvoj YAGO Daily Med MARC Codes List Goodwin Family iServe TCM Gene DIT Semantic Crunch Base RDF Lotico Revyu SW Dog Food (RKB Explorer) OAI ohloh BibBase DBLP (RKB ACM VIVO UF VIVO DBLP (L3S) Explorer) Indiana DBLP RAE2001 VIVO Cornell (FU Berlin) data OS dcs IEEE CiteSeer Project Gutenberg (FUB) ERA STW GESIS Courseware ePrints UN/ LOCODE SIDER STITCH Medi Diseasome OBO Pub Chem KEGG Drug LAAS Budapest IRIT Newcastle IBM Roma dotAC CORDIS KISTI JISC GovTrack rdfabout US SEC Semantic XBRL Linked Sensor Data (Kno.e.sis) rdfabout US Census EUNIS riese Open Cyc Twarql WordNet (VUA) Climbing Linked GeoData WordNet (W3C) Lexvo totl.net Linked Open Numbers UniRef Cornetto Airports Product DB UniSTS Linked CT Taxonomy UniParc Uni Pathway Drug Bank Care Pfam ChEBI PDB Reactome KEGG Cpd HGNC KEGG Pathway UniProt CAS KEGG Enzyme Affymetrix PROSITE ProDom PubMed Gene Ontology SGD Chem2 Bio2RDF GeneID Homolo Gene Gen MGI KEGG Glycan NSF Media KEGG Reaction Geographic Publications User-generated content Government Cross-domain Life sciences Bank OMIM InterPro As of September 2010 September ’10: 203 datasets Linking Open Data cloud diagram, by Richard Cyganiak and Anja Jentzsch. http://lod-cloud.net/
19. RDF Data Volumes . . . . . . are growing – and fast Linked data cloud currently consists of 325 datasets with >25B triples Size almost doubling every year Linked EU Institutions Brazilian Poli- ticians ISTAT Immigration Data Gov.ie Audio LOV User Slideshare tags2con Moseley Scrobbler Feedback 2RDF delicious Bricklink Sussex GTAA Folk (DBTune) Reading St. Magna- Klapp- Lists Andrews tune stuhl- Resource NTU Ox Points Crime Reports UK Hellenic PD Hellenic FBD DB Tropes EUTC Produc- tions Crime (EnAKTing) business data.gov. uk FanHubz John Peel (DBTune) Surge Radio Discogs (Data Incubator) Last.FM Music Brainz (Data Incubator) Music Brainz (DBTune) Music Brainz (zitgist) Lotico RDF ohloh club Linked Crunch- base Jamendo (DBtune) Poké- gnoss Semantic Tweet Ontos News Portal yovisto Manchester Reading Lists Source Code Ecosystem Linked Data Plymouth Reading Lists SSW Thesaur Lists Open Library LinkedL CCN Resource Lists Open Library (Talis) NDL subjects RAMEAU SH Thesau- t4gm info LEM ntnusc Popula- artists pédia Didactal us rus W LIBRIS reegle Ren. Energy Genera- tors EEA Energy (En- AKTing) Open Election Data Project legislation data.gov.uk tion (En- NHS AKTing) (En- AKTing) Mortality (En- AKTing) CO2 Emission (En- AKTing) Ordnance Survey UK Postcodes research data.gov. uk patents data.go v.uk (DBTune) educatio n.data.g ov.uk OpenEI BBC Program mes Openly Local Rechtspraak. nl BBC Wildlife Finder statistics Last.FM (rdfize) Classical (DB Tune) Pokedex Goodwin Family flickr wrappr BBC Music Chronicling America Event Media Linked MDB Portuguese DBpedia Telegraphis New York Times URI Burner Greek DBpedia Open Calais ia semantic web.org Revyu my Experi- ment SW Dog Food RDF Book Mashup theses. fr Sudoc MARC Codes List PSH DDC IdRef Sudoc Calames UB Mannheim data bnf.fr P20 LCSH Rådata nå! ndlna GND VIAF Norwegian MeSH Europeana Ulm Deutsche Bio- graphie lobid Resources NSZL Catalog ECS Wiki lobid Swedish Open Cultural Heritage GovWILD Lichfield Spending ESD standards Scotland Pupils & Exams Traffic Scotland reference data.gov. uk London Gazette CORDIS data.gov.uk intervals TWC LOGD data.gov. uk transport data.gov. uk Eurostat LOIUS NASA (Data Incubator) Eurostat (FUB) Taxon Concept Geo Names World Factbook Freebase Fishes of Texas Geo Species Uberblic Geo Linked Data UMBEL YAGO lingvoj dbpedia lite DBpedia Daily Med iServe OS Project Guten- berg data dcs TCM Gene DIT dataopenac-uk Disea- BNB BibBase ECS Southampton EPrints ECS (RKB Explorer) Southampton OAI DBLP (FU Berlin) DBLP (L3S) DBLP (RKB Explorer) ACM ERA UN/ LOCODE Organisations Budapest Pisa IRIT IBM New- STW GESIS RESEX Scholarometer NVD DEPLOY (RKB some SIDER RAE2001 castle LOCAH CORDIS (FUB) Explorer) GovTrack Eurostat (Ontology Central) Linked Sensor Data (Kno.e.sis) riese Open Cyc Lexvo Enipedia LinkedCT Drug Bank Pfam Eurécom CiteSeer Roma Courseware EURES FTS Linked EDGAR (Ontology Central) US SEC (rdfabout) Scotland Geo- graphy Semantic XBRL GeoWord Net Finnish Municipalities Piedmont Accomodations El Viajero Tourism Italian public schools Turismo de Zaragoza US Census (rdfabout) EUNIS Twarql SMC Journals Climbing Linked GeoData Ocean Drilling Codices AEMET Metoffice Weather Forecasts Janus AMP Yahoo! Geo Planet Weather Stations National Radioactivity totl.net UniProt PDB VIVO Indiana ePrints dotAC IEEE WordNet (VUA) Cornetto LODE Taxono my UniProt (Bio2RDF) PROSITE ProDom HGNC STITCH VIVO Cornell LAAS KISTI RISKS NSF WordNet (W3C) WordNet (RKB Explorer) Alpine Ski Austria GEMET EARTh AGROV OC Open Data Thesaurus SISVU Affymetrix PubMed Gene Ontology Linked Open Colors ChEMBL OMIM InterPro MGI SGD Pub Chem GeneID KEGG Drug VIVO UF JISC KEGG Pathway KEGG Enzyme bible ontology KEGG Reaction ECCOTCP PBAC Media Geographic Publications Airports Product DB Product Types Ontology Italian Museums Smart Link Google Art wrapper UniParc UniRef UniSTS Medi Care UniPath Chem2 Bio2RDF Homolo Gene KEGG Compound KEGG Glycan User-generated content Government Cross-domain JP Sears Open Corporates Amsterdam Museum meducator Linked Open Numbers Reactome OGOLOD way Life sciences As of September 2011 September ’11: 295 datasets Linking Open Data cloud diagram, by Richard Cyganiak and Anja Jentzsch. http://lod-cloud.net/
20. RDF Data Volumes . . . . . . are growing – and fast Linked data cloud currently consists of 325 datasets with >25B triples Size almost doubling every year April ’14: 1091 datasets, ??? triples Max Schmachtenberg, Christian Bizer, and Heiko Paulheim: Adoption of Linked Data Best Practices in Different Topical Domains. In Proc. ISWC, 2014.
21. Outline RDF Introduction gStore: a graph-based SPARQL query engine Answering SPARQL queries using graph pattern matching [Zou et al., PVLDB 2011, VLDB J 2014] gAnswer: Natural Language Question Answering over RDF A Data Driven Approach [Zou et al., SIGMOD 2014; Zheng et al., SIGMOD 2015]
22. Outline RDF Introduction gStore: a graph-based SPARQL query engine Answering SPARQL queries using graph pattern matching [Zou et al., PVLDB 2011, VLDB J 2014] gAnswer: Natural Language Question Answering over RDF A Data Driven Approach [Zou et al., SIGMOD 2014; Zheng et al., SIGMOD 2015]
23. RDF Introduction Everything is an uniquely named resource http://en.wikipedia.org/wiki/Abraham Lincoln
24. RDF Introduction Everything is an uniquely named resource Namespaces can be used to scope the names xmlns:y=http://en.wikipedia.org/wiki y:Abraham Lincoln
25. RDF Introduction Everything is an uniquely named resource Namespaces can be used to scope the names Properties of resources can be defined xmlns:y=http://en.wikipedia.org/wiki y:Abraham Lincoln Abraham Lincoln:hasName “Abraham Lincoln” Abraham Lincoln:BornOnDate: “1809-02-12” Abraham Lincoln:DiedOnDate: “1865-04-15”
26. RDF Introduction Everything is an uniquely named resource Namespaces can be used to scope the names Properties of resources can be defined Relationships with other resources can be defined xmlns:y=http://en.wikipedia.org/wiki y:Abraham Lincoln Abraham Lincoln:hasName “Abraham Lincoln” Abraham Lincoln:BornOnDate: “1809-02-12” Abraham Lincoln:DiedOnDate: “1865-04-15” Abraham Lincoln:DiedIn y:Washington DC
27. RDF Introduction Everything is an uniquely named resource Namespaces can be used to scope the names Properties of resources can be defined Relationships with other resources can be defined Resources can be contributed by different people/groups and can be located anywhere in the web Integrated web “database” xmlns:y=http://en.wikipedia.org/wiki y:Abraham Lincoln Abraham Lincoln:hasName “Abraham Lincoln” Abraham Lincoln:BornOnDate: “1809-02-12” Abraham Lincoln:DiedOnDate: “1865-04-15” Abraham Lincoln:DiedIn y:Washington DC
28. RDF Data Model Triple: Subject, Predicate (Property), Object (s, p, o) Subject: the entity that is described (URI or blank node) Predicate: a feature of the entity (URI) Object: value of the feature (URI, blank node or literal) (s, p, o) ∈ (U ∪ B) × U × (U ∪ B ∪ L) Set of RDF triples is called an RDF graph U Subject Predicate Object UB UB L U: set of URIs B: set of blank nodes L: set of literals Subject Abraham Lincoln Abraham Lincoln Abraham Lincoln Predicate hasName BornOnDate DiedOnDate Object “Abraham Lincoln” “1809-02-12” “1865-04-15”
29. RDF Example Instance URI Prefix: y=http://en.wikipedia.org/wiki Subject Predicate Object y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y: Abraham Lincoln hasName “Abraham Lincoln” y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y: Abraham Lincoln BornOnDate “1809-02-12”’ y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y: Abraham Lincoln DiedOnDate “1865-04-15” y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:Abraham Lincoln bornIn y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:Hodgenville KY y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y: Abraham Lincoln DiedIn y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y: Washington DC y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:Abraham Lincoln title “President” y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:Abraham Lincoln gender “Male” y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y: Washington DC hasName “Washington D.C.” y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:Washington DC foundingYear “1790” y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:Hodgenville KY hasName “Hodgenville” y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:United States hasName “United States” y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:United States hasCapital y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:Washington DC y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:United States foundingYear “1776” y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:Reese Witherspoon bornOnDate “1976-03-22” y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:Reese Witherspoon bornIn y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:New Orleans LA y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:Reese Witherspoon hasName “Reese Witherspoon” y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:Reese Witherspoon gender “Female” y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:Reese Witherspoon title “Actress” y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:New Orleans LA foundingYear “1718” y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:New Orleans LA locatedIn y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:United States y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:Franklin Roosevelt hasName “Franklin D. Roosevelt” y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:Franklin Roosevelt bornIn y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:Hyde Park NY y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:Franklin Roosevelt title “President” y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:Franklin Roosevelt gender “Male” y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:Hyde Park NY foundingYear “1810” y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:Hyde Park NY locatedIn y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:United States y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:Marilyn Monroe gender “Female” y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:Marilyn Monroe hasName “Marilyn Monroe” y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:Marilyn Monroe bornOnDate “1926-07-01” y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:'>y:Marilyn Monroe diedOnDate “1962-08-05” Literal URI
30. RDF Graph “1926-07-01” “Female” bornOnDate gender “1962-08-05” diedOnDate y:Marilyn Monroe hasName “Marilyn Monroe” “Abraham Lincoln” “President” “Male” hasName title gender “1809-02-12” bornOnDate y:Abraham Lincoln bornIn y:Hodgenville KY hasName “Hodgenville” diedOnDate diedIn “Franklin D. Roosevelt” “Male” “1865-04-15” y:Washington D.C. “1776” hasName gender “1976-03-22” foundYear hasName hasCapitalfoundingYear y:Franklin Roosevelt bornOnDate “1790” “Washington D.C.” y:United States title “President” y:Reese Witherspoon bornIn gender title hasName bornIn hasName locatedIn locatedIn “Female” “United States” “Actress”“Reese Witherspoon” y:New Orleans LA y:Hyde Park NY foundingYear foundingYear “1718” “1810”
31. RDF Query Model Query Model - SPARQL Protocol and RDF Query Language Given U (set of URIs), L (set of literals), and V (set of variables), a SPARQL expression is defined recursively: an atomic triple pattern, which is an element of (U ∪ V ) × (U ∪ V ) × (U ∪ V ∪ L) ?x hasName “Abraham Lincoln” P FILTER R, where P is a graph pattern expression and R is a built-in SPARQL condition (i.e., analogous to a SQL predicate) ?x price ?p FILTER(?p < 30) P1 AND/OPT/UNION P2, where P1 and P2 are graph pattern expressions Example: SELECT ?name WHERE { ?m ? c i t y . ?m ?name . ?m ? bd . ? c i t y ‘ ‘ 1 7 1 8 ’ ’ . FILTER ( regex ( s t r ( ? bd ) , ‘ ‘ 1 9 7 6 ’ ’ ) ) }
32. SPARQL Queries SELECT ?name WHERE { ?m ? c i t y . ?m ?name . ?m ? bd . ? c i t y ‘ ‘ 1 7 1 8 ’ ’ . FILTER ( regex ( s t r ( ? bd ) , ‘ ‘ 1 9 7 6 ’ ’ ) ) } FILTER(regex(str (?bd),“1976”)) ?name hasName ?bd bornOnDate ?m bornIn ?city “1718” foundingYear
33. Na¨ıve Triple Store Design SELECT ?name WHERE { ?m ? c i t y . ?m ?name . ?m ? bd . ? c i t y FILTER ( regex ( s t r ( ? bd ) , ‘ ‘ 1 9 7 6 ’ ’ ) ) } Subject Property y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln hasName y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln bornOnDate y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln diedOnDate y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln bornIn y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln diedIn y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln title y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln gender y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC hasName y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC foundingYear y:Hodgenville'>y:Hodgenville KY hasName y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United States hasName y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United States hasCapital y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United States foundingYear y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese Witherspoon bornOnDate y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese Witherspoon bornIn y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese Witherspoon hasName y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese Witherspoon gender y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese Witherspoon title y:New'>y:New'>y:New'>y:New Orleans LA foundingYear y:New'>y:New'>y:New'>y:New Orleans LA locatedIn y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin Roosevelt hasName y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin Roosevelt bornIn y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin Roosevelt title y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin Roosevelt gender y:Hyde'>y:Hyde'>y:Hyde'>y:Hyde Park NY foundingYear y:Hyde'>y:Hyde'>y:Hyde'>y:Hyde Park NY locatedIn y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn Monroe gender y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn Monroe hasName y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn Monroe bornOnDate y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn Monroe diedOnDate Object “Abraham Lincoln” “1809-02-12” “1865-04-15” y:Hodgenville'>y:Hodgenville KY y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC “President” “Male” “Washington D.C.” “1790” “Hodgenville” “United States” y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC “1776” “1976-03-22” y:New'>y:New'>y:New'>y:New Orleans LA “Reese Witherspoon” “Female” “Actress” “1718” y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United States “Franklin D. Roosevelt” y:Hyde'>y:Hyde'>y:Hyde'>y:Hyde Park NY “President” “Male” “1810” y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United States “Female” “Marilyn Monroe” “1926-07-01” “1962-08-05” ‘ ‘1718 ’ ’ .
34. Na¨ıve Triple Store Design SELECT ?name WHERE { ?m ? c i t y . ?m ?name . ?m ? bd . ? c i t y ‘ ‘ 1 7 1 8 ’ ’ . FILTER ( regex ( s t r ( ? bd ) , ‘ ‘ 1 9 7 6 ’ ’ ) ) } Subject Property y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln hasName y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln bornOnDate y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln diedOnDate y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln bornIn y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln diedIn y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln title y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln gender y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC hasName y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC foundingYear y:Hodgenville'>y:Hodgenville KY hasName y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United States hasName y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United States hasCapital y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United States foundingYear y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese Witherspoon bornOnDate y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese Witherspoon bornIn y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese Witherspoon hasName y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese Witherspoon gender y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese Witherspoon title y:New'>y:New'>y:New'>y:New Orleans LA foundingYear y:New'>y:New'>y:New'>y:New Orleans LA locatedIn y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin Roosevelt hasName y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin Roosevelt bornIn y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin Roosevelt title y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin Roosevelt gender y:Hyde'>y:Hyde'>y:Hyde'>y:Hyde Park NY foundingYear y:Hyde'>y:Hyde'>y:Hyde'>y:Hyde Park NY locatedIn y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn Monroe gender y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn Monroe hasName y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn Monroe bornOnDate y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn Monroe diedOnDate Object “Abraham Lincoln” “1809-02-12” “1865-04-15” y:Hodgenville'>y:Hodgenville KY y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC “President” “Male” “Washington D.C.” “1790” “Hodgenville” “United States” y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC “1776” “1976-03-22” y:New'>y:New'>y:New'>y:New Orleans LA “Reese Witherspoon” “Female” “Actress” “1718” y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United States “Franklin D. Roosevelt” y:Hyde'>y:Hyde'>y:Hyde'>y:Hyde Park NY “President” “Male” “1810” y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United States “Female” “Marilyn Monroe” “1926-07-01” “1962-08-05” SELECT T2 . o b j e c t FROM T a s T1 , T a s T2 , T a s T3 , T a s T4 WHERE T1 . p r o p e r t y=” b o r n I n ” AND T2 . p r o p e r t y=” hasName ” AND T3 . p r o p e r t y=” bornOnDate ” AND T1 . s u b j e c t=T2 . s u b j e c t AND T2 . s u b j e c t=T3 . s u b j e c t AND T4 . p r o p e t y=” f o u n d i n g Y e a r ” AND T1 . o b j e c t=T4 . s u b j e c t AND T4 . o b j e c t=” 1718 ” AND T3 . o b j e c t LIKE ’ %1976% ’
35. Na¨ıve Triple Store Design SELECT ?name WHERE { ?m ? c i t y . ?m ?name . ?m ? bd . ? c i t y ‘ ‘ 1 7 1 8 ’ ’ . FILTER ( regex ( s t r ( ? bd ) , ‘ ‘ 1 9 7 6 ’ ’ ) ) } Subject Property y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln hasName y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln bornOnDate y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln diedOnDate y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln bornIn y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln diedIn y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln title y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln gender y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC hasName y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC foundingYear y:Hodgenville'>y:Hodgenville KY hasName y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United States hasName y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United States hasCapital y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United States foundingYear y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese Witherspoon bornOnDate y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese Witherspoon bornIn y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese Witherspoon hasName y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese Witherspoon gender y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese'>y:Reese Witherspoon title y:New'>y:New'>y:New'>y:New Orleans LA foundingYear y:New'>y:New'>y:New'>y:New Orleans LA locatedIn y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin Roosevelt hasName y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin Roosevelt bornIn y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin Roosevelt title y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin'>y:Franklin Roosevelt gender y:Hyde'>y:Hyde'>y:Hyde'>y:Hyde Park NY foundingYear y:Hyde'>y:Hyde'>y:Hyde'>y:Hyde Park NY locatedIn y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn Monroe gender y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn Monroe hasName y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn Monroe bornOnDate y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn'>y:Marilyn Monroe diedOnDate Object “Abraham Lincoln” “1809-02-12” “1865-04-15” y:Hodgenville'>y:Hodgenville KY y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC “President” “Male” “Washington D.C.” “1790” “Hodgenville” “United States” y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC “1776” “1976-03-22” y:New'>y:New'>y:New'>y:New Orleans LA “Reese Witherspoon” “Female” “Actress” “1718” y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United States “Franklin D. Roosevelt” y:Hyde'>y:Hyde'>y:Hyde'>y:Hyde Park NY “President” “Male” “1810” y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United'>y:United States “Female” “Marilyn Monroe” “1926-07-01” “1962-08-05” Too many self-joins! SELECT T2 . o b j e c t FROM T a s T1 , T a s T2 , T a s T3 , T a s T4 WHERE T1 . p r o p e r t y=” b o r n I n ” AND T2 . p r o p e r t y=” hasName ” AND T3 . p r o p e r t y=” bornOnDate ” AND T1 . s u b j e c t=T2 . s u b j e c t AND T2 . s u b j e c t=T3 . s u b j e c t AND T4 . p r o p e t y=” f o u n d i n g Y e a r ” AND T1 . o b j e c t=T4 . s u b j e c t AND T4 . o b j e c t=” 1718 ” AND T3 . o b j e c t LIKE ’ %1976% ’
36. Existing Solutions 1. Property table Each class of objects go to a different table ⇒ similar to normalized relations Eliminates some of the joins 2. Vertically partitioned tables For each property, build a two-column table, containing both subject and object, ordered by subjects Can use merge join (faster) Good for subject-subject joins but does not help with subject-object joins 3. Exhaustive indexing Create indexes for each permutation of the three columns Query components become range queries over individual relations with merge-join to combine Excessive space usage
37. Property Tables Grouping by entities; Jena [Wilkinson et al., SWDB 03] ,FlexTable [Wang et al., DASFAA 10] , DB2-RDF [Bornea et al., SIGMOD 13] Clustered property table:'>table: group together the properties that tend to occur in the same (or similar) subjects Property-class table:'>table: cluster the subjects with the same type of property into one property table Subject Property Object y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln hasName “Abraham Lincoln” y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln bornOnDate “1809-02-12” y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln diedOnDate “1865-04-15” y:Washington'>y:Washington'>y:Washington'>y:Washington DC hasName “Washington D.C.” y:Washington'>y:Washington'>y:Washington'>y:Washington DC foundingYear “1790” ... ... ... Subject hasName bornOnDate diedOnDate y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln “Abraham Lincoln” 1809-02-12 1865-04-15 y:Reese Witherspoon “Reese Witherspoon” 1976-03-22 Subject hasName foundingYear y:Washington'>y:Washington'>y:Washington'>y:Washington DC “Washington D.C.” 1790 y:Hyde Park NY “Hyde Park” 1810’
38. Property Tables Grouping by entities; Jena [Wilkinson et al., SWDB 03] ,FlexTable [Wang et al., DASFAA 10] , DB2-RDF [Bornea et al., SIGMOD 13] Clustered property table:'>table: group together the properties that Advatnetnadgetso occur in the same (or similar) subjects FPerowpeerrjtoyi-ncslass table:'>table: cluster the subjects with the same type of property into one property table If the data is structured, we have a relational system – similar to norSmubajelcitzed relatPiroonpesrty Object y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln hasName “Abraham Lincoln” y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln bornOnDate “1809-02-12” y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln diedOnDate “1865-04-15” y:Washington'>y:Washington'>y:Washington'>y:Washington DC hasName “Washington D.C.” y:Washington'>y:Washington'>y:Washington'>y:Washington DC foundingYear “1790” ... ... ... Subject hasName bornOnDate diedOnDate y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln “Abraham Lincoln” 1809-02-12 1865-04-15 y:Reese Witherspoon “Reese Witherspoon” 1976-03-22 Subject hasName foundingYear y:Washington'>y:Washington'>y:Washington'>y:Washington DC “Washington D.C.” 1790 y:Hyde Park NY “Hyde Park” 1810’
39. Property Tables Grouping by entities; Jena [Wilkinson et al., SWDB 03] ,FlexTable [Wang et al., DASFAA 10] , DB2-RDF [Bornea et al., SIGMOD 13] Clustered property table:'>table: group together the properties that Advatnetnadgetso occur in the same (or similar) subjects FPerowpeerrjtoyi-ncslass table:'>table: cluster the subjects with the same type of property into one property table If the data is structured, we have a relational system – similar to norSmubajelcitzed relatPiroonpesrty Object y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln hasName “Abraham Lincoln” y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln bornOnDate “1809-02-12” Disadvantay:gAebrsaham Lincoln diedOnDate “1865-04-15” y:Washington'>y:Washington DC hasName “Washington D.C.” Potenyt:iWaalslhyingatolnoDtCoffouNndUinLgYLeasr “1790” ... ... ... Clustering is not trivial Multi-valued properties are complicated Subject hasName bornOnDate diedOnDate y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln “Abraham Lincoln” 1809-02-12 1865-04-15 y:Reese Witherspoon “Reese Witherspoon” 1976-03-22 Subject hasName foundingYear y:Washington'>y:Washington DC “Washington D.C.” 1790 y:Hyde Park NY “Hyde Park” 1810’
40. Binary Tables Grouping by properties: For each property, build a two-column table, containing both subject and object, ordered by subjects [Abadi et al., VLDB 07] Also called vertical partitioned tables n two column tables (n is the number of unique properties in the data) Subject Property Object y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln hasName “Abraham Lincoln” y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln bornOnDate “1809-02-12” y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln diedOnDate “1865-04-15” y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC hasName “Washington D.C.” y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC foundingYear “1790” ... ... ... hasName Subject Object y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln “Abraham Lincoln” y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC “Washington D.C.” bornOnDate Subject Object y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln 1809-02-12 y:Reese Witherspoon 1976-03-22 foundingYear Subject Object y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC 1790 y:Hyde Park NY 1810
41. Binary Tables Grouping by properties: For each property, build a two-column table, containing both subject and object, ordered by subjects Adva[nAtbaagdeiset al., VLDB 07] ASulspopcoartllsedmuveltrit-ivcaallupedartpirtoiopneerdtietsables nNotwNoUcLoLlusmn tables (n is the number of unique properties in tNhoe cdlautsate)ring Read only nSeuebjdecetd attribuPrtoepsert(yi.e. Olbejsecst I/O) y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln hasName “Abraham Lincoln” Good perfory:mAbaranhacme LfinocrolnsubobrnjeOcnDt-asteub“1j8e0c9t-02j-o12in” s y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln diedOnDate “1865-04-15” y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC hasName “Washington D.C.” y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC foundingYear “1790” ... ... ... hasName Subject Object y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln “Abraham Lincoln” y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC “Washington D.C.” bornOnDate Subject Object y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln 1809-02-12 y:Reese Witherspoon 1976-03-22 foundingYear Subject Object y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC 1790 y:Hyde Park NY 1810
42. Binary Tables Grouping by properties: For each property, build a two-column table, containing both subject and object, ordered by subjects Adva[nAtbaagdeiset al., VLDB 07] ASulspopcoartllsedmuveltrit-ivcaallupedartpirtoiopneerdtietsables nNotwNoUcLoLlusmn tables (n is the number of unique properties in tNhoe cdlautsate)ring Read only nSeuebjdecetd attribuPrtoepsert(yi.e. Olbejsecst I/O) y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln hasName “Abraham Lincoln” Good perfory:mAbaranhacme LfinocrolnsubobrnjeOcnDt-asteub“1j8e0c9t-02j-o12in” s y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln diedOnDate “1865-04-15” y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC hasName “Washington D.C.” Disadvantages y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC foundingYear “1790” ... ... ... Not useful for subject-object joins Expensive inserts hasName bornOnDate Subject Object y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln “Abraham Lincoln” y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC “Washington D.C.” Subject Object y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln 1809-02-12 y:Reese Witherspoon 1976-03-22 foundingYear Subject Object y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington'>y:Washington DC 1790 y:Hyde Park NY 1810
43. Exhaustive Indexing RDF-3X [Neumann and Weikum, PVLDB 08] , Hexastore [Weiss et al., PVLDB 08] Strings are mapped to ids using a mapping table Original triple table Subject Property Object y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln hasName “Abraham Lincoln” y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln bornOnDate “1809-02-12” y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln diedOnDate “1865-04-15” y:Washington'>y:Washington'>y:Washington'>y:Washington DC hasName “Washington D.C.” y:Washington'>y:Washington'>y:Washington'>y:Washington DC foundingYear “1790” Encoded triple table Subject Property Object 0 1 2 0 3 4 0 5 6 7 1 8 7 9 10 Mapping table ID Value 0 y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham'>y:Abraham Lincoln 1 hasName 2 “Abraham Lincoln” 3 bornOnDate 4 “1809-02-12” 5 diedOnDate 6 “1865-04-15” 7 y:Washington'>y:Washington'>y:Washington'>y:Washington DC 8 “Washington D.C.” 9 foundingYear 10 “1790”
44. Exhaustive Indexing RDF-3X [Neumann and Weikum, PVLDB 08] , Hexastore [Weiss et al., PVLDB 08] Strings are mapped to ids using a mapping table Triples are indexed in a clustered B+ tree in lexicographic order Subject 0 0 0 7 7 Property 1 3 5 1 9 Object 2 4 6 8 10 B+ tree Easy querying through mapping table
45. Exhaustive Indexing RDF-3X [Neumann and Weikum, PVLDB 08] , Hexastore [Weiss et al., PVLDB 08] Strings are mapped to ids using a mapping table Triples are indexed in a clustered B+ tree in lexicographic order Create indexes for permutations of the three columns: SPO, SOP, PSO, POS, OPS, OSP Subject 0 0 0 7 7 Property 1 3 5 1 9 Object 2 4 6 8 10 B+ tree Easy querying through mapping table
46. Exhaustive Indexing–Query Execution Each triple pattern can be answered by a range query Joins between triple patterns computed using merge join Join order is easy due to extensive indexing Subject 0 0 0 7 7 ... Property 1 3 5 1 9 ... Object 2 4 6 8 10 ... ID Value 0 y:Abraham Lincoln 1 hasName 2 “Abraham Lincoln” 3 bornOnDate 4 “1809-02-12” 5 diedOnDate 6 “1865-04-15” 7 y:Washington DC 8 “Washington D.C.” 9 foundingYear 10 “1790”
47. Exhaustive Indexing–Query Execution Each triple pattern can be answered by a range query Joins between triple patterns computed using merge join Join order is easy due to extensive indexing AdvSaunbtjaegcets Property Object ID Value E0liminates so1me of the2joins – the0y becyo:mAberarahnagme Lqiunecroilens M0erge join is3easy and 4fast 0 5 6 1 hasName 2 “Abraham Lincoln” 7 1 8 3 bornOnDate 7 9 10 4 “1809-02-12” ... ... ... 5 diedOnDate 6 “1865-04-15” 7 y:Washington DC 8 “Washington D.C.” 9 foundingYear 10 “1790”
48. Exhaustive Indexing–Query Execution Each triple pattern can be answered by a range query Joins between triple patterns computed using merge join Join order is easy due to extensive indexing AdvSaunbtjaegcets Property Object ID Value E0liminates so1me of the2joins – the0y becyo:mAberarahnagme Lqiunecroilens M0erge join is3easy and 4fast 0 5 6 1 hasName 2 “Abraham Lincoln” Disadv7antages 1 8 3 bornOnDate Sp7... ace usage 9 ... 10 ... 4 “1809-02-12” 5 diedOnDate 6 “1865-04-15” 7 y:Washington DC 8 “Washington D.C.” 9 foundingYear 10 “1790”
49. Outline RDF Introduction gStore: a graph-based SPARQL query engine Answering SPARQL queries using graph pattern matching [Zou et al., PVLDB 2011, VLDB J 2014] gAnswer: Natural Language Question Answering over RDF A Data Driven Approach [Zou et al., SIGMOD 2014; Zheng et al., SIGMOD 2015]
50. gStore – General Idea We work directly on the RDF graph and the SPARQL query graph Answering SPARQL query ≡ subgraph matching Subgraph matching is computationally expensive Use a signature-based encoding of each entity and class vertex to speed up matching Filter-and-evaluate Use a false positive algorithm to prune nodes and obtain a set of candidates; then do more detailed evaluation on those We develop an index (VS∗-tree) over the data signature graph (has light maintenance load) for efficient pruning
51. 0. Start with RDF Graph G FILTER(regex(str (?bd),“1976”)) ?name hasName ?bd bornOnDate ?m bornIn ?city “1718” foundingYear “1926-07-01” “Female” bornOnDate gender “1962-08-05” diedOnDate y:Marilyn Monroe hasName “Marilyn Monroe” “Abraham Lincoln” “President” “Male” hasName title gender “1809-02-12” bornOnDate y:Abraham Lincoln bornIn y:Hodgenville KY hasName “Hodgenville” diedOnDate diedIn “Franklin D. Roosevelt” “Male” “1865-04-15” y:Washington D.C. “1776” hasName gender “1976-03-22” foundYear hasName hasCapitalfoundingYear y:Franklin Roosevelt bornOnDate “1790” “Washington D.C.” y:United States title “President” y:Reese Witherspoon bornIn gender title hasName bornIn hasName locatedIn locatedIn “Female” “United States” “Actress”“Reese Witherspoon” y:New Orleans LA y:Hyde Park NY foundingYear foundingYear “1718” “1810”
52. 0. Start with RDF Graph G FILTER(regex(str (?bd),“1976”)) ?name hasName ?bd bornOnDate ?m bornIn ?city “1718” foundingYear Finding matches over a large graph is not a trivial task! “1926-07-01” “Female” bornOnDate gender “1962-08-05” diedOnDate y:Marilyn Monroe hasName “Marilyn Monroe” “Abraham Lincoln” “President” “Male” hasName title gender “1809-02-12” bornOnDate y:Abraham Lincoln bornIn y:Hodgenville KY hasName “Hodgenville” diedOnDate diedIn “Franklin D. Roosevelt” “Male” “1865-04-15” y:Washington D.C. “1776” hasName gender “1976-03-22” foundYear hasName hasCapitalfoundingYear y:Franklin Roosevelt bornOnDate “1790” “Washington D.C.” y:United States title “President” y:Reese Witherspoon bornIn gender title hasName bornIn hasName locatedIn locatedIn “Female” “United States” “Actress”“Reese Witherspoon” y:New Orleans LA y:Hyde Park NY foundingYear foundingYear “1718” “1810”
53. gStore System Architecture Offline Input RDF data RDF Parser RDF Triples RDF Graph Builder RDF Graph Encoding Module Signature Graph VS*-tree builder VS*-tree Storage VS*-tree Store Online Input Results SPARQL Query SPARQL Parser Query Graph Encoding Module Signature Graph Filter Module Node Candidate Join Module Key-Value Store
54. gStore
55. gStore
56. Peer Review Comments Charu Aggarwal, ACM Fellow, IBM T. J. Watson Researcher ——NeMa: Fast Graph Search with Label Similarity, Proc. of VLDB: 181-192 (2013)
57. Outline RDF Introduction gStore: a graph-based SPARQL query engine Answering SPARQL queries using graph pattern matching [Zou et al., PVLDB 2011, VLDB J 2014] gAnswer: Natural Language Question Answering over RDF A Data Driven Approach [Zou et al., SIGMOD 2014; Zheng et al., SIGMOD 2015]
58. gAnswer: Natural Language Question Answering Over Knowledge Graph–A Graph Data Driven Approach An Easy-to-Use Interface to Access Knowledge Graph It is interesting to both academia and industry. Interdisciplinary research between database and NLP (natural language processing) communities.
59. gAnswer: Natural Language Question Answering Over Knowledge Graph–A Graph Data Driven Approach An Easy-to-Use Interface to Access Knowledge Graph It is interesting to both academia and industry. Interdisciplinary research between database and NLP (natural language processing) communities. gAnswer
60. Running Example Question: Who was married to an actor that play in Philadelphia ? Subject Antonio Banderas Antonio Banderas Antonio Banderas Philadelphia (film) Jonathan Demme Philadelphia Aaron McKie James Anderson Constantin Stanislavski Philadelphia 76ers An Actor Prepares Property type spouse starring type director type bornIn playForTeam create type type Object actor Melanie Griffith Philadelphia (film) film film city Philadelphia Philadelphia 76ers An Actor Prepares Basketball team Book
61. Running Example Question: Who was married to an actor that play in Philadelphia ? Subject Antonio Banderas Antonio Banderas Antonio Banderas Philadelphia (film) Jonathan Demme Philadelphia Aaron McKie James Anderson Constantin Stanislavski Philadelphia 76ers An Actor Prepares Property type spouse starring type director type bornIn playForTeam create type type Object actor Melanie Griffith Philadelphia (film) film film city Philadelphia Philadelphia 76ers An Actor Prepares Basketball team Book Melanie Griffith
62. Existing Solutions Who was married to an actor that play in Philadelphia ? SELECT ? y Translate NL Question to structured queries WHERE { ?x starring Philadelphia ( film ). ?x type actor . ?x spouse ?y . } Query Processing Melanie Griffith
63. Existing Solutions Ambiguity Who was married to an actor that play in Philadelphia ? SELECT ? y WHERE { Translate NL Question to structured queries ?x starring Philadelphia ( film ). Philadelpha ?x type actor . ?x spouse ?y . } Philadelpha (film) Query Processing Philadelpha 76ers Melanie Griffith
64. Existing Solutions Ambiguity Who was married to an actor that play in Philadelphia? SELECT ? y WHERE { Translate NL Question to structured queries ?x starring Philadelphia ( film ). playForTeam ?x type actor . ?x spouse ?y . } starring Query Processing director Melanie Griffith
65. Our Method: Motivation–Data Driven actor film city type type Antonio Banderas type Philadelphia starring spouse Philadelphia (film) bornIn Melanie Griffith director Aaron McKie Jonathan Demme James Anderson Basketball team PlayForTeam type Philadelphia 76ers Constantin Stanislavski create Book type An Actor Prepares
66. Our Method: Motivation–Data Driven actor film city type type Antonio Banderas type Philadelphia starring spouse Philadelphia (film) bornIn Melanie Griffith director Aaron McKie Jonathan Demme Who was married to an actor that play in Philadelphia ? “Who” “Philadelphia” “be married to” “play in” “actor”/”that” James Anderson Basketball team PlayForTeam type Philadelphia 76ers Constantin Stanislavski create Book type An Actor Prepares
67. Our Method: Motivation–Data Driven actor film city type type Antonio Banderas type Philadelphia starring spouse Philadelphia (film) bornIn Melanie Griffith director Aaron McKie Jonathan Demme Who was married to an actor that play in Philadelphia ? (director, 1.0) (starring, 0.9) (spouse, 1.0)(playForTeam, 1.0) “Who” “Philadelphia” “be married to” “play in” “actor”/”that” James Anderson Basketball team PlayForTeam type Philadelphia 76ers Constantin Stanislavski create Book type An Actor Prepares ?Who (actor, 1.0) (Philadelphia, 1.0) (An Actor Prepares, 0.9) (Philadelphia (film), 0.9) (Philadelphia 76ers, 0.8)
68. Our Method: Motivation–Data Driven actor film city type type Antonio Banderas type Philadelphia starring spouse Philadelphia (film) bornIn Melanie Griffith director Aaron McKie Jonathan Demme Who was married to an actor that play in Philadelphia ? (director, 1.0) (starring, 0.9) (spouse, 1.0)(playForTeam, 1.0) “Who” “Philadelphia” “be married to” “play in” “actor”/”that” James Anderson Basketball team PlayForTeam type Philadelphia 76ers Constantin Stanislavski create Book type An Actor Prepares ?Who (actor, 1.0) (Philadelphia, 1.0) Combine Disambiguation (AnaAncdtoQr PuererpyarTeso,g0e.t9h) er ! (Philadelphia (film), 0.9) (Philadelphia 76ers, 0.8)
69. Experiments: Datasets RDF repository: DBPedia Table : Statistics of RDF Graph DBpedia Number of Entities 5.2 million Number of Triples 60 million Number of Predicates 1643 Size of RDF Graphs (in GB) 6.1 Relation Phrase Dictionary: Patty Table : Statistics of Relation Phrase Dataset Number of Textual Patterns Number of Entity Pairs Average Entity Pair Number For Each Pattern wordnet-wikipedia 350,568 3,862,304 11 freebase-wikipedia 1,631,530 15,802,947 9
70. Experiments: Online Benchmark: QALD-3, 99 Natural Language Questions Table : Evaluating QALD-3 Testing Questions (on DBpedia) Processed Right Partially Recall Precision F-1 Our 76 32 11 0.40 0.40 0.40 Method squall2sparql 96 77 13 0.85 0.89 0.87 CASIA 52 29 8 0.36 0.35 0.36 Scalewelis 70 1 38 0.33 0.33 0.33 RTV 55 30 4 0.34 0.32 0.33 Intui2 99 28 4 0.32 0.32 0.32 SWIP 21 14 2 0.15 0.16 0.16 DEANNA 27 21 0 0.21 0.21 0.21
71. Experiments: Online ID Q2 Q3 Q14 Q17 Q19 Q20 Q21 Q22 Q24 Q27 Q28 Q30 Q35 Q39 Q41 Q42 Q44 Q45 Q54 Q58 Q63 Q70 Q74 Q76 Q77 Q81 Q83 Q84 Q86 Q89 Q98 Q100 Questions Who was the successor of John F. Kennedy? Who is the mayor of Berlin? Give me all members of Prodigy? Give me all cars that are produced in Germany ? Give me all people that were born in Vienna and died in Berlin ? How tall is Michael Jordan ? What is the capital of Canada ? Who is the governor of Wyoming ? Who was the father of Queen Elizabeth II? Sean Parnell is the governor of which U.S. state ? Give me all movies directed by Francis Ford Coppola. What is the birth name of Angela Merkel ? Who developed Minecraft ?. Give me all companies in Munich. Who founded Intel? Who is the husband of Amanda Palmer ? Which cities does the Weser flow through ? Which countries are connected by the Rhine ? What are the nicknames of San Francisco ? What is the time zone of Salt Lake City ? Give me all Argentine films. Is Michelle Obama the wife of Barack Obama ? When did Michael Jackson die ? List the children of Margaret Thatcher. Who was called Scarface? Which books by Kerouac were published by Viking Press ? How high is the Mount Everest ? Who created the comic Captain America ? What is the largest city in Australia ? In which city was the former Dutch queen Juliana buried ? Which country does the creator of Miffy come from ? Who produces Orangina ? Response Time (in ms) 1699 677 811 297 2557 942 1342 796 538 1210 577 250 2565 1312 1105 1418 1139 736 321 316 427 316 258 1139 719 796 635 589 1419 1700 2121 367
72. Experiments: Online Running Time (in ms) 100000 10000 1000 100 10 Q2 Q20 Q21 Q22 Q28 Q35 Q41 Q42 Q44 Q45 Q54 Q74 Q76 Q83 Q84 Q86 Question Understanding in DEANNA Overall time in DEANNA Question Understanding in Our System Overall time in Our System Figure : Online Running Time Comparison
73. Experiments: Online Figure : QALD-4 Results
74. An Example: using gAnswer+gStore in CDBLP
75. An Example: using gAnswer+gStore in CDBLP
76. An Example: using gAnswer+gStore in CDBLP
77. Conclusions Graph Database is a Possible Way for RDF Knowledge Base Management.
78. Conclusions Graph Database is a Possible Way for RDF Knowledge Base Management. Subgraph Matching is a Strong Tool.
79. Conclusions Graph Database is a Possible Way for RDF Knowledge Base Management. Subgraph Matching is a Strong Tool. Using RDF repository, how to Provide Knowledge Services for Applications and Common Users?
80. Thank you! gStore gAnswer