social network analysis, semantic web and learning networks

40
SNA & Semantic Web (and LN) Rory Sie

Upload: rory-sie

Post on 06-May-2015

1.034 views

Category:

Education


1 download

DESCRIPTION

Session 2 of the Learning Networks Social Networks Seminar. It presents a recap of SNA terms, and introduces the Semantic Web and how it could be applied to Learning Networks.

TRANSCRIPT

Page 1: Social Network Analysis, Semantic Web and Learning Networks

SNA & Semantic Web (and LN)

Rory Sie

Page 2: Social Network Analysis, Semantic Web and Learning Networks

Outline

•Recap

•Semantic Web

•Use for LN

Page 3: Social Network Analysis, Semantic Web and Learning Networks

Recap

Page 4: Social Network Analysis, Semantic Web and Learning Networks

Network measures

•network: density, connectivity, centralization

•community: factions, cliques

•individual: betweenness, degree, closeness

Page 5: Social Network Analysis, Semantic Web and Learning Networks

Network measures

•network: density, connectivity, centralization

•community: factions, cliques

•individual: betweenness, degree, closeness

Page 6: Social Network Analysis, Semantic Web and Learning Networks

Network measures

•network: density, connectivity, centralization

•community: factions, cliques

•individual: betweenness, degree, closeness

Page 7: Social Network Analysis, Semantic Web and Learning Networks

Data storage

• Adjacency matrix (R, UCINET)

• GML/XGMML (Cytoscape, Gephi)

• Pajek Network (Pajek, UCINET)

Page 8: Social Network Analysis, Semantic Web and Learning Networks

Data storage

• Adjacency matrix (R, UCINET)

• GML/XGMML (Cytoscape, Gephi)

• Pajek Network (Pajek, UCINET)

<?xml version="1.0" encoding="UTF-8" standalone="yes"?><graph label="PLN for ID " directed="1"><node id="n26" label="n26"><att type="string" name="PeerName" value="Rory Sie"/></node><node id="n27" label="n27"><att type="string" name="PeerName" value="Adriana Berlanga"/></node><edge id="e0" label="e0" source="n26" target="n27"><att type="string" name="interaction" value="colleague""/>

</edge></graph>

Page 9: Social Network Analysis, Semantic Web and Learning Networks

Data storage

• Adjacency matrix (R, UCINET)

• GML/XGMML (Cytoscape, Gephi)

• Pajek Network (Pajek, UCINET)

Page 10: Social Network Analysis, Semantic Web and Learning Networks

Analysis

But what if you want to do this real-time / online?

Page 11: Social Network Analysis, Semantic Web and Learning Networks

CytoscapeWeb

•http://cytoscapeweb.cytoscape.org

•Cytoscape, but online

•Great for visualization

Page 12: Social Network Analysis, Semantic Web and Learning Networks

Connect R to web

RemoteREngine package

Page 13: Social Network Analysis, Semantic Web and Learning Networks
Page 14: Social Network Analysis, Semantic Web and Learning Networks

Web 1.0

Page 15: Social Network Analysis, Semantic Web and Learning Networks

Web 2.0

Page 16: Social Network Analysis, Semantic Web and Learning Networks

Semantic Web (3.0)

writeswrites

about place

wri

tes

about

resource

Page 17: Social Network Analysis, Semantic Web and Learning Networks

Semantic Web (3.0)

writes

writes

about place

wri

tes

about

resource

Page 18: Social Network Analysis, Semantic Web and Learning Networks

Semantic Web (3.0)

writes

writes

about place

wri

tes

about

resource

learn

s fr

om

friend o

f

mother of

follows

Page 19: Social Network Analysis, Semantic Web and Learning Networks

learning

netw

orks

Page 20: Social Network Analysis, Semantic Web and Learning Networks

Knowledge Representation

•RDF

•Triple store (e.g. Sesame)

•Query language (e.g. SPARQL)

Page 21: Social Network Analysis, Semantic Web and Learning Networks

Knowledge Representation

•RDF

•Triple store (e.g. Sesame)

•Query language (e.g. SPARQL)

“Rory”“learns from”“Adriana”

Page 22: Social Network Analysis, Semantic Web and Learning Networks

Knowledge Representation

•RDF

•Triple store (e.g. Sesame)

•Query language (e.g. SPARQL)

“Rory”“learns from”“Adriana”subject predicate object

Page 23: Social Network Analysis, Semantic Web and Learning Networks

Knowledge Representation

•RDF

•Triple store (e.g. Sesame)

•Query language (e.g. SPARQL)

subject predicate objecttriple“Rory”“learns from”“Adriana”

Page 25: Social Network Analysis, Semantic Web and Learning Networks

Knowledge Representation

•RDF

•Triple store (e.g. Sesame)

•Query language (e.g. SPARQL)

Page 26: Social Network Analysis, Semantic Web and Learning Networks

htt

p:/

/ww

w.a

g-n

bi.de/r

ese

arc

h/s

wrl

engin

e/

Page 27: Social Network Analysis, Semantic Web and Learning Networks

Knowledge Representation

•RDF

•Triple store (e.g. Sesame)

•Query language (e.g. SPARQL)

Page 29: Social Network Analysis, Semantic Web and Learning Networks

SPARQL

SELECT ?tutor

WHERE

{

<http://ln.org/person/Rory> <http://ln.org/learns_from> ?tutor

}

Page 30: Social Network Analysis, Semantic Web and Learning Networks

Result

<http://ln.org/person/Adriana>

Page 31: Social Network Analysis, Semantic Web and Learning Networks

How can this help us?

•store learning networks data in RDF

•use SNA to analyse network, individuals, communities, topics

Page 32: Social Network Analysis, Semantic Web and Learning Networks

CSCL script and roles

Capuano et al, 2011)

Page 33: Social Network Analysis, Semantic Web and Learning Networks

SemWeb, LNs and SNA

peer learner

peer learner

friend

father

mother

adapted from Ereteo

degree = 5

Page 34: Social Network Analysis, Semantic Web and Learning Networks

SemWeb, LNs and SNA

peer learner

peer learner

friend

father

mother

adapted from Ereteo

degree<family> = 2

Page 35: Social Network Analysis, Semantic Web and Learning Networks

SemWeb, LNs and SNA

peer learner

peer learner

friend

father

mother

adapted from Ereteo

degree<friend> = 1

Page 36: Social Network Analysis, Semantic Web and Learning Networks

SemWeb, LNs and SNA

peer learner

peer learner

friend

father

mother

adapted from Ereteo

degree<peer learner> = 2

Page 37: Social Network Analysis, Semantic Web and Learning Networks

SPARQL n-degreeselect ?y count(?x) as ?degree where{{?x $path ?yfilter(match($path, star(param[type])))filter(pathLength($path) <= param[length]) } UNION{?y $path ?xfilter(match($path, star(param[type]))) filter(pathLength($path) <= param[length]) }} group by ?y

Page 38: Social Network Analysis, Semantic Web and Learning Networks

Summary

Semantic Web and Social Network Analysis help us make sense of different types of data that are in a social network

Page 39: Social Network Analysis, Semantic Web and Learning Networks

[email protected]

http://www.open.ou.nl/rse

openrory, maisonpoublon

Rory Sie

openrse

http://nl.linkedin.com/in/rorysie

thebigbangrory.blogspot.com

Page 40: Social Network Analysis, Semantic Web and Learning Networks

References

• R project (http://www.r-project.org/)

• UCINET (https://sites.google.com/site/ucinetsoftware/home)

• Gephi (http://gephi.org/)

• Cytoscape (http://www.cytoscape.org)

• Capuano, N., Laria, G., Mazzoni, E., Pierri, A., & Mangione, G. R. (2011). Improving Role Taking in CSCL Script Using SNA and Semantic Web. 2011 IEEE 11th International Conference on Advanced Learning Technologies, 636-637. Ieee. doi:10.1109/ICALT.2011.197

• Berners-lee, B. T., Hendler, J., & Lassila, O. (2001). The Semantic Web. Scientific American.

• Guillaume Ereteo’s PhD defense (http://www.slideshare.net/ereteog/phd-defense-semantic-social-network-analysis)

• Microformats (http://microformats.org/)