Swap2010 twitter minining using semantic web technologies and linked data
- 1. @twitter Mining #MicroblogsUsing #SemanticTechnologiesSelver Softic, Martin Ebner, Herbert Mühlburger , Thomas Altmann, Behnam Taraghi
- 2. Web 2.0 - well knownstoryWeb 2.0 technologiesbroughtuserscloserto Web …Wikis, Blogs, Forums …Podcasts, RSS, XML …… thenusersstartedtogeneratecontent …Source: http:mediabistro.com
- 3. From Web toSocial WebResult = a vastofinformationText, Pictures, Audio, Videos ….Communication, networking, exchangeofdataWeb becamemore personalCultural, geographicalandsocialbordersdisappearedSource: http://www.ignitesocialmedia.com
- 7. But still disconnected ?source: www.pidgintech.com
- 8. Data is still captured in Walled Garden!
- 9. StatementsSocial Web relies on usersandcommunicationamongthemWhilecommunicatingusersproduceorconsumecontentSocialsitesaredatasilosrich on varietyofinformationThisinformationcouldbeinterestingfor:monitoring of trends, advertising, statistics, reputation, news broadcasting , tagging …Thisdataiscaptured in Walledgarden !!!
- 10. QuestionsHowtousethisdatatogainmoreusefulinsightsWhataretheadvantagesof online (offline) search on such dataandhowtoreachit in an uniform wayIs itpossibletostructurize, connectandexposethedata in order tobeusedbyhumansandmachinesmoreefficientlyWhatwould an architecturelooklikeforthisissue
- 11. Social Web TrendsMicrobloggingSocialBookmarkingSocial NetworkingSocial MarketingSharing Photos, Videos …Source: http://socialwebresearch.com
- 12. MicroblogsMicroblogsUsedforcommunication,publishingandinformationexchangeSimple forprocessingInformation generatedbymany different usersSocialuserrelationsTripartitecommunicationstructureVarietyofinformationsNoboundariesbyculture,locationortechnology (mobile users)TwitterMost PopularLarge amountoddataBut limitedAccording: http://an.kaist.ac.kr/traces/WWW2010.html41.7 million user profiles, 1.47 billion social relations, 4,262 trending topics, and 106 million tweets
- 13. SemanticaspectsandTwitterTwitterUser realtionsTweetsasshortinformationartefactsCommunication withtripartitepatternTime relatedinformationVocabulariesSIOC, FOAF, Dublin Core
- 14. Linked Data andTwitterTwittercontainsinfos on:People, Organisations, Locations, Trends …LOD CloudcontainsBillionsoftriplesabout:Geolocations , dataaboutscience, government, commonknowledge, persons, news …VocabulariesMOAT, CommmonTag
- 17. GrabeeterSearch in your TweetsFilter your Tweets by dateSearch in your Tweets offline using the Grabeeter ClientFilter your tweets offline using the Grabeeter ClientGrabeeter provides an API
- 18. Triplification Module AuthorDateContentReciever<tweet url="http://grabeeter.tugraz.at/tweet/199272" text="Sitting in Prater #vienna, launch party. Nice" screen_name="selvers" created="2010-08-19" twitterUrl="http://twitter.com/selvers/status/21606926237"/>RDF StoreTriplifier
- 19. Triplification Module@prefix foaf: <http://xmlns.com/foaf/0.1/> .@prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> .@prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> .@prefix sioc: <http://rdfs.org/sioc/ns#> .@prefix sioct: <http://rdfs.org/sioc/types#> .@prefix dcterms: <http://purl.org/dc/terms/#> .<http://twitter.com/selvers/status/21606926237> rdf:typesioct:MicroblogPost ;sioc:content "Sitting in Prater #vienna, launch party. Nice" ;sioc:has_creator <http://twitter.com/selvers/> ;foaf:maker <http://grabeteer.tugraz.at/foaf/selvers/> ;dcterms:created “2010-08-19” ;rdfs:sameAs <http://grabeeter.tugraz.at/tweet/199272> .<http://twitter.com/selvers/> rdf:typefoaf:Person ;foaf:name "SelverSoftic" ;foaf:depiction <http://a0.twimg.com/profile_images/905118560/f9e4b6eba.13070201_3_normal.jpg> ;foaf:knows <http://twitter.com/hmuehlburger/> ;foaf:knows <http://twitter.com/mhausenblas/> ;foaf:knows <http://twitter.com/mebner/> . …
- 20. Interlinking ModuleHashtags (People, Organisation, Locations)MOAT, CommonTagLater NLP processedcontent, SILK FrameworkSELECT ?post ?content ?maker ?nameWHERE {?post rdf:typesioct:MicroblogPost;foaf:maker ?maker; ?makerfoaf:name ?name;sioc:content ?content.FILTER(regex(?content,#vienna))} Classifiertag: tagName "vienna" ;moat: tagMeaning<http://dbpedia .org/resource/Vienna>tag: taggedResource <http://twitter.com/selvers/status/2160692623>
- 21. Analysis
- 22. Conclusions & OutlookCurrentstateofthearttechnologiessufficetorealisetheproposedarchitectureparadigmInterlinkingwith LOD Cloud (Tweet-O-Sphere)Involving NLP MethodsSentiment classification(Re)TaggingofTweetsProviding SPARQL Endpoint + Lookup ServiceasresearchinterfaceSocialSemantic Web Apps
- 23. Questions?