@twitter Mining #MicroblogsUsing #SemanticTechnologies<br />Selver Softic, Martin Ebner, Herbert Mühlburger , Thomas Altma...
Web 2.0 -  well knownstory<br />Web 2.0 technologiesbroughtuserscloserto Web …<br />Wikis, Blogs, Forums …<br />Podcasts, ...
From Web toSocial Web<br />Result = a vastofinformation<br />Text, Pictures, Audio, Videos ….<br />Communication, networki...
Social Media Boom!<br />
Socialsitesaredatasilos<br />source: www.pidgintech.com<br />
But still disconnected ?<br />source: www.pidgintech.com<br />
Data is still captured in Walled Garden!<br />
Statements<br />Social Web relies on usersandcommunicationamongthem<br />Whilecommunicatingusersproduceorconsumecontent<br...
Questions<br />Howtousethisdatatogainmoreusefulinsights<br />Whataretheadvantagesof online (offline) search on such dataan...
Social Web Trends<br />Microblogging<br />SocialBookmarking<br />Social Networking<br />Social Marketing<br />Sharing Phot...
Microblogs<br />Microblogs<br />Usedforcommunication,publishingandinformationexchange<br />Simple forprocessing<br />Infor...
SemanticaspectsandTwitter<br />Twitter<br />User realtions<br />Tweetsasshortinformationartefacts<br />Communication witht...
Linked Data andTwitter<br />Twittercontainsinfos on:<br />People, Organisations, Locations, Trends …<br />LOD Cloudcontain...
Architecture model<br />
Acquisition - Grabeeter<br />
Grabeeter<br />Search in your Tweets<br />Filter your Tweets by date<br />Search in your Tweets offline using the Grabeete...
Triplification Module <br />Author<br />Date<br />Content<br />Reciever<br /><tweet url="http://grabeeter.tugraz.at/tweet/...
Triplification Module<br />@prefix foaf: <http://xmlns.com/foaf/0.1/> .<br />@prefix rdfs: <http://www.w3.org/2000/01/rdf-...
Interlinking Module<br />Hashtags (People, Organisation, Locations)<br />MOAT, CommonTag<br />Later NLP processedcontent, ...
Analysis<br />
Conclusions & Outlook<br />Currentstateofthearttechnologiessufficetorealisetheproposedarchitectureparadigm<br />Interlinki...
Questions?<br />
Upcoming SlideShare
Loading in...5
×

@twitter Mining #Microblogs Using #Semantic Technologies

2,182

Published on

Presenation of Selver Softic at 6th Workshop on Semantic Web Applications and Perspectives (SWAP 2010)

Published in: Education
0 Comments
1 Like
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total Views
2,182
On Slideshare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
19
Comments
0
Likes
1
Embeds 0
No embeds

No notes for slide

@twitter Mining #Microblogs Using #Semantic Technologies

  1. 1. @twitter Mining #MicroblogsUsing #SemanticTechnologies<br />Selver Softic, Martin Ebner, Herbert Mühlburger , Thomas Altmann, Behnam Taraghi<br />
  2. 2. Web 2.0 - well knownstory<br />Web 2.0 technologiesbroughtuserscloserto Web …<br />Wikis, Blogs, Forums …<br />Podcasts, RSS, XML …<br />… thenusersstarted<br />togeneratecontent …<br />Source: http:mediabistro.com<br />
  3. 3. From Web toSocial Web<br />Result = a vastofinformation<br />Text, Pictures, Audio, Videos ….<br />Communication, networking, exchangeofdata<br />Web becamemore personal<br />Cultural, geographicalandsocialbordersdisappeared<br />Source: http://www.ignitesocialmedia.com<br />
  4. 4. Social Media Boom!<br />
  5. 5.
  6. 6. Socialsitesaredatasilos<br />source: www.pidgintech.com<br />
  7. 7. But still disconnected ?<br />source: www.pidgintech.com<br />
  8. 8. Data is still captured in Walled Garden!<br />
  9. 9. Statements<br />Social Web relies on usersandcommunicationamongthem<br />Whilecommunicatingusersproduceorconsumecontent<br />Socialsitesaredatasilosrich on varietyofinformation<br />Thisinformationcouldbeinterestingfor:<br />monitoring of trends, advertising, statistics, reputation, news broadcasting , tagging …<br />Thisdataiscaptured in Walledgarden !!!<br />
  10. 10. Questions<br />Howtousethisdatatogainmoreusefulinsights<br />Whataretheadvantagesof online (offline) search on such dataandhowtoreachit in an uniform way<br />Is itpossibletostructurize, connectandexposethedata in order tobeusedbyhumansandmachinesmoreefficiently<br />Whatwould an architecturelooklikeforthisissue<br />
  11. 11. Social Web Trends<br />Microblogging<br />SocialBookmarking<br />Social Networking<br />Social Marketing<br />Sharing Photos, Videos …<br />Source: http://socialwebresearch.com<br />
  12. 12. Microblogs<br />Microblogs<br />Usedforcommunication,publishingandinformationexchange<br />Simple forprocessing<br />Information generatedbymany different users<br />Socialuserrelations<br />Tripartitecommunicationstructure<br />Varietyofinformations<br />Noboundariesbyculture,locationortechnology (mobile users)<br />Twitter<br />Most Popular<br />Large amountoddata<br />But limited<br />According: http://an.kaist.ac.kr/traces/WWW2010.html<br />41.7 million user profiles, 1.47 billion social relations, 4,262 trending topics, and 106 million tweets<br />
  13. 13. SemanticaspectsandTwitter<br />Twitter<br />User realtions<br />Tweetsasshortinformationartefacts<br />Communication withtripartitepattern<br />Time relatedinformation<br />Vocabularies<br />SIOC, FOAF, Dublin Core<br />
  14. 14. Linked Data andTwitter<br />Twittercontainsinfos on:<br />People, Organisations, Locations, Trends …<br />LOD Cloudcontains<br />Billionsoftriplesabout:<br />Geolocations , dataaboutscience, government, commonknowledge, persons, news …<br />Vocabularies<br />MOAT, CommmonTag<br />
  15. 15. Architecture model<br />
  16. 16. Acquisition - Grabeeter<br />
  17. 17. Grabeeter<br />Search in your Tweets<br />Filter your Tweets by date<br />Search in your Tweets offline using the Grabeeter Client<br />Filter your tweets offline using the Grabeeter Client<br />Grabeeter provides an API<br />
  18. 18. Triplification Module <br />Author<br />Date<br />Content<br />Reciever<br /><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"/><br />RDF <br />Store<br />Triplifier<br />
  19. 19. Triplification Module<br />@prefix foaf: <http://xmlns.com/foaf/0.1/> .<br />@prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> .<br />@prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> .<br />@prefix sioc: <http://rdfs.org/sioc/ns#> .<br />@prefix sioct: <http://rdfs.org/sioc/types#> .<br />@prefix dcterms: <http://purl.org/dc/terms/#> .<br /><http://twitter.com/selvers/status/21606926237> rdf:typesioct:MicroblogPost ;<br />sioc:content "Sitting in Prater #vienna, launch party. Nice" ;<br />sioc:has_creator <http://twitter.com/selvers/> ;<br />foaf:maker <http://grabeteer.tugraz.at/foaf/selvers/> ;<br />dcterms:created “2010-08-19” ;<br />rdfs:sameAs <http://grabeeter.tugraz.at/tweet/199272> .<br /><http://twitter.com/selvers/> rdf:typefoaf:Person ;<br />foaf:name "SelverSoftic" ;<br />foaf:depiction <http://a0.twimg.com/profile_images/905118560/f9e4b6eba.13070201_3_normal.jpg> ;<br />foaf:knows <http://twitter.com/hmuehlburger/> ;<br />foaf:knows <http://twitter.com/mhausenblas/> ;<br />foaf:knows <http://twitter.com/mebner/> . <br />…<br />
  20. 20. Interlinking Module<br />Hashtags (People, Organisation, Locations)<br />MOAT, CommonTag<br />Later NLP processedcontent, SILK Framework<br />SELECT ?post ?content ?maker ?name<br />WHERE {<br />?post rdf:typesioct:MicroblogPost;<br />foaf:maker ?maker;<br /> ?makerfoaf:name ?name;<br />sioc:content ?content.<br />FILTER(regex(?content,#vienna))<br />} <br />Classifier<br />tag: tagName "vienna" ;<br />moat: tagMeaning<br /><http://dbpedia .org/resource/Vienna><br />tag: taggedResource <http://twitter.com/selvers/status/2160692623><br />
  21. 21. Analysis<br />
  22. 22. Conclusions & Outlook<br />Currentstateofthearttechnologiessufficetorealisetheproposedarchitectureparadigm<br />Interlinkingwith LOD Cloud (Tweet-O-Sphere)<br />Involving NLP Methods<br />Sentiment classification<br />(Re)TaggingofTweets<br />Providing SPARQL Endpoint + Lookup Serviceasresearchinterface<br />SocialSemantic Web Apps<br />
  23. 23. Questions?<br />
  1. A particular slide catching your eye?

    Clipping is a handy way to collect important slides you want to go back to later.

×