knowledgestore
play

KnowledgeStore Scalable Framework for Interlinking Text and - PowerPoint PPT Presentation

Mar 19, 2024 •256 likes •707 views

KnowledgeStore Scalable Framework for Interlinking Text and Knowledge Marco Rospocher, Bernardo Magnini, Luciano Serafini Fondazione Bruno Kessler (FBK) INTRODUCTION Information is typically available both in unstructured and

  1. KnowledgeStore — Scalable Framework for Interlinking Text and Knowledge — Marco Rospocher, Bernardo Magnini, Luciano Serafini Fondazione Bruno Kessler (FBK)

  2. INTRODUCTION • Information is typically available both in unstructured and structured form • Deep and Large scale NLP now enables to bridge the two “world” • Development of frameworks for integrating unstructured and structured content only partially investigated

  3. KnowledgeStore • A scalable, fault-tolerant, and Semantic Web grounded storage system to jointly store, manage, retrieve, and query, both structured and unstructured data

  4. KnowledgeStore Motivating scenario • Among a collection of news articles, a user is interested in retrieving all 2014 articles reporting statements of a 20th century US president where he is positively mentioned as “commander-in-chief”.

  5. KnowledgeStore Motivating scenario • Among a collection of news articles, a user is interested in retrieving all 2014 articles reporting statements of a 20th century US president where he is positively mentioned as “commander-in-chief”.

  6. KnowledgeStore In a nutshell

  7. KnowledgeStore Exploitation • Enhanced applications (e.g., decision support systems) • Developing, debugging, training, and evaluating NLP and knowledge processing tasks • Reasoning on Extracted Information (e.g., on Events) • Text Exploration

  8. OUTLINE • A walk through the KnowledgeStore • The KnowledgeStore “live” • The KnowledgeStore in NewsReader • Next Challenges

  9. KnowledgeStore Functional View

  10. KnowledgeStore Data Model • It’s Flexible • It’s an OWL 2 Ontology

  11. KnowledgeStore Example: Data Model For Event Extraction from News

  12. KnowledgeStore Architectural View Any application Java applications (HTTP access to the KS, possibly KnowledgeStore Java client exploiting SPARQL client libraries) Client-side SPARQL endpoint CRUD endpoint Server-side KnowledgeStore Frontend Server Virtuoso OMID Zookeeper Hadoop HDFS HBase (single node) (single node) ( multiple server nodes) (mult. nodes) (name & data nodes) distributed transaction Resource Mention Context Representation Entity Axiom synchronization manager

  13. KnowledgeStore Looking through the glass box • The KnowledgeStore User Interface • lookup: given the URI of an object (i.e., resource, mention, entity), retrieves all the content about it • SPARQL: run arbitrary queries against the SPARQL endpoint

  14. KnowledgeStore recording history by processing massive streams of daily news ICT 316404 FP7-ICT-2011-8 www.newsreader-project.eu Jan 2013 - Dec 2015

  15. KnowledgeStore in

  16. KnowledgeStore LIVE DEMO Dataset used in the Demo (NewsReader Project): • Domain: Global Automotive Industry • 1.3M news documents (2003-2013), provided by LexisNexis (www.lexisnexis.nl) • 1.3M NLP annotation files (NAF format) obtained processing the news (NewsReader NLP Pipeline) • 205M mentions of events , persons , organisations , locations , time expressions ... • 535M of RDF triples about events , persons , organisations , locations , time expressions ...: • 439M extracted from text • 96M coming from selected background knowledge (DBpedia)

  17. KnowledgeStore in

  18. KnowledgeStore in Decision Making on top of the KnowledgeStore

  19. KnowledgeStore in Exploited in Three Hack Day Events

  20. KnowledgeStore in Exploited in Three Hack Day Events • Capable of handling: • large number of requests (>110K) • multiple concurrent requests (40 requests/sec.) • low response time (30-214ms)

  21. KnowledgeStore Reasoning on Events • Inferring Knowledge not Explicitly Mentioned in Text (powered by Event Situation Ontology - ESO) • Example: “Yesterday, Chrysler hired Jim Press to lead its sales and marketing” 2005 2006 2007 2008 2009 2010

  22. KnowledgeStore Reasoning on Events • Inferring Knowledge not Explicitly Mentioned in Text (powered by Event Situation Ontology - ESO) • Example: “Yesterday, Chrysler hired Jim Press to lead its sales and marketing” At time: 2007/09/07 Even ent: hire Even ent typ ype: e: ESO:JoiningAnOrganization Even ent roles: roles: ESO:employer dbpedia:Chrysler ESO:employee dbpedia:Jim_Press 2005 2006 2007 2008 2009 2010

  23. KnowledgeStore Reasoning on Events • Inferring Knowledge not Explicitly Mentioned in Text (powered by Event Situation Ontology - ESO) • Example: “Yesterday, Chrysler hired Jim Press to lead its sales and marketing” At time: 2007/09/07 Even ent: hire Even ent typ ype: e: ESO:JoiningAnOrganization Even ent roles: roles: ESO:employer dbpedia:Chrysler ESO:employee dbpedia:Jim_Press pre-Situation dbpedia:Jim_Press ESO:notEmployedAt dbpedia:Chrysler 2005 2006 2007 2008 2009 2010

  24. KnowledgeStore Reasoning on Events • Inferring Knowledge not Explicitly Mentioned in Text (powered by Event Situation Ontology - ESO) • Example: “Yesterday, Chrysler hired Jim Press to lead its sales and marketing” At time: 2007/09/07 Even ent: hire Even ent typ ype: e: ESO:JoiningAnOrganization Even ent roles: roles: ESO:employer dbpedia:Chrysler ESO:employee dbpedia:Jim_Press pre-Situation dbpedia:Jim_Press ESO:notEmployedAt dbpedia:Chrysler post-Situation dbpedia:Jim_Press ESO:employedAt dbpedia:Chrysler 2005 2006 2007 2008 2009 2010

  25. KnowledgeStore Reasoning on Events • Applied on the 1.3M global automotive industry news • Extremely fast: 1,333s ( ~22m ) to process ~500M triples • Triggered 2M new triples (i.e., not in the text ), organised in 397,885 situations • 255,470 events have at least a pre/post/during situation: • 71.2% of the events having at least two distinct roles

  26. KnowledgeStore Looking ahead to the future…

  27. KnowledgeStore Beyond “Asserted” Knowledge… • What knowledge can be additionally inferred from what mentioned in text? 2007 2008 2009 2010 2011 2012

  28. KnowledgeStore Beyond “Asserted” Knowledge… • What knowledge can be additionally inferred from what mentioned in text? At time: 2007/09/07 Even ent: hire Even ent typ ype: e: ESO:JoiningAnOrganization Even ent roles: roles: ESO:employer B ESO:employee A pre-Situation post-Situation A ESO:employedAt B A ESO:notEmployedAt B 2007 2008 2009 2010 2011 2012

  29. KnowledgeStore Beyond “Asserted” Knowledge… • What knowledge can be additionally inferred from what mentioned in text? At time: 2007/09/07 At time: 2011/08/05 Even ent: hire Even ent: fire Even ent typ ype: e: ESO:JoiningAnOrganization Even ent typ ype: e: ESO:LeavingAnOrganization Even ent roles: roles: ESO:employer B Even ent roles: roles: ESO:employer C ESO:employee A ESO:employee A pre-Situation post-Situation A ESO:employedAt B A ESO:notEmployedAt B post-Situation pre-Situation A ESO:notEmployedAt C A ESO:employedAt C 2007 2008 2009 2010 2011 2012

  30. KnowledgeStore Beyond “Asserted” Knowledge… • What knowledge can be additionally inferred from what mentioned in text? At time: 2007/09/07 At time: 2011/08/05 Even ent: hire Even ent: fire Even ent typ ype: e: ESO:JoiningAnOrganization Even ent typ ype: e: ESO:LeavingAnOrganization Even ent roles: roles: ESO:employer B Even ent roles: roles: ESO:employer C ESO:employee A ESO:employee A pre-Situation post-Situation A ESO:employedAt B A ESO:notEmployedAt B post-Situation pre-Situation A ESO:notEmployedAt C A ESO:employedAt C 2007 2008 2009 2010 2011 2012

  31. KnowledgeStore Beyond “Asserted” Knowledge… • What knowledge can be additionally inferred from what mentioned in text? At time: 2007/09/07 At time: 2011/08/05 Even ent: hire Even ent: fire Even ent typ ype: e: ESO:JoiningAnOrganization Even ent typ ype: e: ESO:LeavingAnOrganization Even ent roles: roles: ESO:employer B Even ent roles: roles: ESO:employer C ESO:employee A ESO:employee A pre-Situation post-Situation A ESO:employedAt B A ESO:notEmployedAt B post-Situation pre-Situation A ESO:notEmployedAt C A ESO:employedAt C 2007 2008 2009 2010 2011 2012 Even ent: fire Even ent typ ype: e: ESO:LeavingAnOrganization Even ent roles: roles: ESO:employer B ESO:employee A

  32. KnowledgeStore Beyond “Asserted” Knowledge… • What knowledge can be additionally inferred from what mentioned in text? At time: 2007/09/07 At time: 2011/08/05 Even ent: hire Even ent: fire Even ent typ ype: e: ESO:JoiningAnOrganization Even ent typ ype: e: ESO:LeavingAnOrganization Even ent roles: roles: ESO:employer B Even ent roles: roles: ESO:employer C ESO:employee A ESO:employee A pre-Situation post-Situation A ESO:employedAt B A ESO:notEmployedAt B post-Situation pre-Situation A ESO:notEmployedAt C A ESO:employedAt C 2007 2008 2009 2010 2011 2012 Even ent: fire Even ent typ ype: e: ESO:LeavingAnOrganization Even ent roles: roles: ESO:employer B Even ent: resign ESO:employee A Even ent typ ype: e: ESO:LeavingAnOrganization Even ent roles: roles: ESO:employer B ESO:employee A

More recommend