Social Interaction Ontology

Post on 16-May-2015

12.072 views 0 download

Tags:

description

Research report for social interaction ontology - including activity streams, online presence project, SIOC and FOAF.

Transcript of Social Interaction Ontology

Social Semantic Weband Social Interaction Ontology

Seokchan (Channy) Yun

Biomedical Knowledge Engineering LaboratorySeoul National University

June 8th 2009

Research Draft

Agenda• Online Social Network

– Traditional research

• Social Semantic Web– FOAF/SIOC– Tripartite Social Ontology– Social Semantic Network

• Activity-based Approach– Online Presence Project– Activity Streams

• Social Interaction Ontology– Research Plan

Social Networks in Socialogy• Social Network used firstly

J.A. Barnes (1954)• Research methodology in

Social Sciences• Generally sameAs Network

(Graph) Theory– Milgram’s six degrees of

separation theory– Erdős number– Dunbar’s Number – 150

(average 124)– Diffusion of innovations

Representation of Social Network

• Graph Model with Node and Edge

Emerging Online Social Network• New opportunities for social science

– Explicit and implicit social network information– Large scale data sets– Dynamic data– Different modalities (profiles, email, IM, Twitter…)

• Challenges– Theoretical– Friend on the Web = Friend in reality?– Extracting information– Heterogeneity– Quality of data– Time and space complexity– Ethical and legal challenges

• Semantic technologies can help with some of the technical challenges

History

• First Mover– Classmates.com,

Match.com and sixdegree.com

– Friendster and Orkut

• Majority– Myspace and Facebook– Linkedin.com– Twitter.com

Classification

How succeed?• Allows a user to create and maintain an online network of

close friends or business associates for social and professional reasons:– Friendships and relationships– Offline meetings– Curiosity about others– Business opportunities– Job hunting

• Allows a user to share interests based on object-centered sociality with meaning– Sharing photo, video and bookmark– Life streaming over SNS– Broadcasting and publishing of my own content

Status of Online SNS

John Breslin, The Social Semantic Web: An Introduction (2009)

Social Semantic Information Spaces

FOAF

Example of FOAF

<foaf:Person><foaf:name>

Seokchan Yun</foaf:name><foaf:mbox_sha1sum>

241021fb0e6289f92815fc210f9e9137262c252e</foaf:mbox_sha1sum>

<foaf:homepage rdf:resource="http://www.creation.net" /><foaf:knows><foaf:Person>

<foaf:name>Hong-Gee Kim</foaf:name><foaf:mbox rdf:resource=”mailto:hgkim@snu.ac.kr”/><rdfs:seeAlso rdf:resource=“http://.../foaf.rdf />

</foaf:Person></foaf:Person>

Common Structure

SIOC (John Breslin)

Example of SIOC

<sioc:Post rdf:about="http://koreacrunch.com/archive/firefox-supports-geolocation-service"><dcterms:title>Firefox supports Geolocation Service</dcterms:title><dcterms:created>2006-09-07T09:33:30Z</dcterms:created><sioc:has_container rdf:resource="http://www.creation.net/?sioc_type=site#weblog"/><sioc:has_creator>

<sioc:User rdf:about="http://www.creation.net/" rdfs:label="Cloud"></sioc:User>

</sioc:has_creator><sioc:content>Firefox 3.5 Beta 4 and higher starts to support W3C’s Geolocation APIs

developed by Google.com based...</sioc:content><sioc:topic rdfs:label="Firefox" rdf:resource="http://koreacrunch.com/tags/firefox/"/><sioc:topic rdfs:label="Geolocation" rdf:resource="http://koreacrunch.com/tags/..."/><sioc:has_reply>

<sioc:Post rdf:about="http://koreacrunch.com/archive/351#comment-123928"></sioc:Post>

</sioc:has_reply></sioc:Post>

16

FOAF+ SIOC

17

FOAF+SIOC+SKOS

skos:isSubjectOfsioc:topic

Disconnected sites on the Social Web

Tripartite Social Ontology (Peter Mica)

• A graph model of ontologies based on tripartite graphs of actors, concepts and instances– Actors: users– Concepts: tags– Instances: objects

• Emergent semantics– General idea: observe semantics in the way agents interact

(use concepts)• Bottom-up ontologies

• Semantics = syntax + statistics

Dimensional Ontology Oci and Oac

e.g. Delicious• When looking at co-occurrence of terms (Oci )

– Network reflects language use– Better for clustering, determining ambiguity of terms and

finding synonyms• E.g. travel - cote, provence, villa, azur, mas, holiday, vacation,

tourism, france, heritage

• When looking at community overlaps (Oac )– Network reflects the domain– Better for finding broader/narrower terms, non-trivial

relationships• E.g. google – gmail, picasa, youtube…

e.g. Flink

Social Semantic Network (Jason Jung)

Remained Question?

What’s Definition of Online Friends?

Online Friend != RealFOAF’s knows is not knowing!

Well-known Friends 9%

Colleagues 7%

Meet once in offline 25%

Knowing only name 12%

Famous person 3%

Unknown FoaF 13%

Everyone requested 32%

Known

Unknown

http://answers.polldaddy.com/poll/1230119/?view=results

Twitter

Facebook

me2DAY

LinkedIn

Challenges

• Discovering Knowledge

• Degree of User relationship– Coupling between users (high vs. weak)– Removing coupling bias in analysis

Abdullah Al Reshood roommateOf Aafia SiddiqueAafia Siddique isMemberOf Al QuedaAafia Siddique traveledTo PakistanMounir-al Motassad traveledTo Pakistan

Abdullah Al Reshood isRelatedTo Al Queda ?Al Queda isRelatedTo Pakistan ?

Online Presence Project (Milan Stankovic)

• Feel of Presense– Status Messages– Online Status (Busy, Available, Away…)– Current listening music, activities…

Activity Streams (Chris Messina)

• Lightweight simple Atom based syndication for user’s activities

• Widely supported by Facebook, MySpace etc.• Basic Format

– User, Verb, Noun

Example of Activity Streams<entry>

<id>tag:photopanic.example.com,2008:activity01</id><title>Geraldine posted a Photo on PhotoPanic</title><published>2008-11-02T15:29:00Z</published><link rel="alternate" type="text/html“ href="/geraldine/activities/1" /><activity:verb>

http://activitystrea.ms/schema/1.0/post</activity:verb><activity:object>

<id>tag:photopanic.example.com,2008:photo01</id><title>My Cat</title><published>2008-11-02T15:29:00Z</published><link rel="alternate" type="text/html“ href="/geraldine/photos/1" /><activity:object-type>

tag:atomactivity.example.com,2008:photo</activity:object-type><source>

<title>Geraldine's Photos</title><link rel="self" type="application/atom+xml“ href="/geraldine/photofeed.xml" /><link rel="alternate" type="text/html“ href="/geraldine/" />

</source></activity:object><content type="html">

&lt;p&gt;Geraldine posted a Photo on PhotoPanic&lt;/p&gt;&lt;img src="/geraldine/photo1.jpg"&gt;

</content></entry>

Other approaches

Questions• Twitter

– There are many spammers and followers.– Whom I should follow? Who is expert?

• me2DAY– There are many friends– Who disconnected in my friendship?

• Flickr– There are many photos.– What’s good photos enjoying with friend?

• RateMDs– There are many doctors.– What’s good doctors recommended by friends?

Answers

• 1. Measuring degree of user relationship– Twitter: Following < RT < Reply < Direct Message < SMS– me2Day: me2 < Link < Reply < Memo < Gift < SMS

– Exporting Social Interaction Ontology

• 2. Modeling of user degree– Similarity formula (A,B)

• 3. Integration data for answer

Pre-work for solving problem

• Social web evolves direct sharing and broadcasting instead of document link based distribution and knowledge discovering. – Social Interaction is more important in social networks.– FriendFeed, Facebook life streaming, Twitter

• Need to represent “Degree between people”– Writing simple ontology represents interaction

• Channy replies Hong-Gee (What) (When) in Facebook

• John retweets Channy (What) (When) in Twitter

Why new ontology?• SIOC

– Only focusing on relationship with site(forum), contents and person.

• OPO– Only focusing “Presence” not to be interested in “Activity”– No description on various interaction such as Twitter’s RT– Need to more practical

• Activity streams– Atom (RDF) based– Only description for Person / Verb / Object

Twitter interaction

Facebook interaction

me2DAY interaction

Social Interaction Ontology• Focusing on User-to-User

– User / Verb / User in Sites• E.g. fromAccount Act toAccount

• Connection with FOAF and SIOC– foaf:Person– foaf:holdsAccount

• sioc:User rdfs:subClassOf foaf:onlineAccount– interAct:fromAccount rdfs:sioc:User

• Expression for various interaction– Verb : comment, reply, me2, RT, hashtag…

• E.g act:comment sioc:Sites

select count(?y) as ?cdegree{{ ?y interAct:retweet ?x}UNION{?x interAct:directmessage ?y}

}group by ?x

Research Plan• Who disconnected in my friendship on me2DAY?

– Modeling Social Interactive Ontology– Export me2DAY activity to SIO– Measuring coupling-degree index

• Distance = # of interaction/ time interval

• Priority = normalized value for each interactions

– Evaluation with user’s reaction for alert

• Whom I should follow? Who is expert in Twitter?– Export twitter activity to SIO– Measuring coupling-degree index– Evaluation with user’s reaction for recommendation

Interaction OntologyExporter

Interaction OntologyExporter

Interaction OntologyExporter

Q&A

channy@snu.ac.krhttp://www.creation.net

Twitter: @channyun