Upload
marc-smith
View
7.349
Download
0
Embed Size (px)
DESCRIPTION
An overview of the Network Overview Discovery and Exploration add-in for Excel 2007 (NodeXL), a social network analysis add-in for the familiar spreadsheet application. Visualize twitter, flickr, facebook, and email networks with just a few mouse clicks.
Citation preview
SOCIAL MEDIA NETWORK ANALYSIS WITH NODEXL
Network Overview, Discovery and Exploration for Excel 2007
Marc A. SmithChief Social ScientistConnected Action Consulting Group
[email protected]://www.ConnectedAction.nethttp://www.codeplex.com/nodexlhttp://www.twitter.com/marc_smithhttp://delicious.com/marc_smith/Paper http://www.flickr.com/photos/marc_smithhttp://www.facebook.com/marc.smith.sociologisthttp://www.linkedin.com/in/marcasmithhttp://www.slideshare.net/Marc_A_Smith
Dr. Marc A. Smith
• Research on the social network structure of social media, online community, collaboration.
• Currently working on:
– NodeXL: Social Network Analysishttp://www.codeplex.com/nodexl
– Mapping link structure of social media
– Reporting tools for community and brands
– Discovering social roles in social media, building an ecological model of computer-mediated collective action
Ph.D. in Sociology, UCLA, 2002Chief Social Scientist, Connected Action Consulting Group, Silicon Valley, CAhttp://[email protected]
Problem: No network chart in Excel
Problem: No network chart in
Excel
NodeXL: Network analysis and visualization tool
• Cyclic Graph data structures have limited support in existing Office tools
• Network analysis is of growing importance in academic, commercial, and Internet social media contexts
• Existing network analysis tools have command line interfaces or demand steep learning curves
• Many network data sets already live in Excel!
NodeXL: Goal: Make SNA easier
• Existing Social Network Tools are challenging for many novice users
• Tools like Excel are widely used• Leveraging a spreadsheet as a host for SNA
lowers barriers to network data analysis and display
Social Network Analysis ToolkitTools to support the study of the social network structure of
social media and other directed graph structures
Sociologist
“What are the structures of
communication in scientific
discussions?”
User Experience Information Visualization
“What are the best UI/UX workflows for network
analysis tools?”
Computer Scientist Algorithmicist for Social
Network Measures
“What are the measures and algorithms needed for understanding networks?”
10
The Ties that Blind?
11
Reply-To NetworkNetwork at distance 2 for the most prolific author of the microsoft.public.internetexplorer.general newsgroup
The Ties that Blind?
Darwin Bell13
Pajek without modification can sometimes reveal structures of great interest.
The Ties that Blind?
Two “answer people” with an emerging 3rd.
Mapping Newsgroup Social Ties
Microsoft.public.windowsxp.server.general15
16
• Answer person– Outward ties to local isolates– Relative absence of triangles– Few intense ties
• Reply Magnet– Ties from local isolates often
inward only– Sparse, few triangles– Few intense ties
Distinguishing attributes:
17
Distinguishing attributes:• Answer person– Outward ties to local isolates– Relative absence of triangles– Few intense ties
• Discussion person– Ties from local isolates often
inward only– Dense, many triangles– Numerous intense ties
18
Clear and consistent signaturesof an “Answer Person”
• Light touch to numerous threads initiated by someone else
• Most ties are outward to local isolates• Many more ties to small fish than big fish
1
10
100
0 1 2 4 8 16 32 64
19
Roles Project• Using Netscan
data to derive social roles in Usenet
• Next steps: quantify & explore in more depth
Answer Person, microsoft.public.windows.server.general
Discussion, rec.kites
Flame, alt.flame
Social Support, alt.support.divorce
PUBLISHED in HICSS, JCMC, JoSS, IEEE Internet Communications (special issue on Social Networks) 20
Relationships among #OCS09 mentioning Twitter users
Relationships among “NodeXL” Mentioning Twitter users
NodeXL Imports User and Keyword Networks from Twitter
Dr. Robert Ackland• Works at the intersection of
empirical social science and computer science, developing new approaches and related tools (involving information retrieval, data visualisation and social network analysis) for studying online social and organisational networks
• Established Virtual Observatory for the Study of Online Networks (VOSON) project in 2005 - http://voson.anu.edu.au
• Teaches two masters courses: Social Science of the Internet” and Online Research Methods”
• Fellow and Masters Coordinator in Australian Demographic and Social Research Institute, at the Australian National University
• Degrees in economics from Univ. of Melbourne, Yale Univ. (as Fulbright Scholar) and ANU, where he completed his PhD
Growing NodeXL user community: nearing 18,000 downloads
Network of relationships among “Honda” mentioning Twitter users
Network of relationships among “Opendata” mentioning Twitter users
Network of relationships among “SharePoint” mentioning Twitter users
NodeXL Calculates a number of Network Metrics and “subgraphs”
Sub graph images of each “ego-network”
Network of relationships among “Intuit” mentioning Twitter users
NodeXL supports a number of graph layouts
Harel-Koren
Fruchterman-Reingold
Harel-Koren layout of Marc Smith Facebook Egonetwork
Fruchterman-Reingold layout of Marc Smith Facebook Egonetwork
FR layout of the relationships among #WIN09 Mentioning Twitter users
US Senate votingrelationships
US Senate votingrelationships
Changed party in 2009
Systematic Yet Flexible Network Analysis Tasks1. Overall network metrics (e.g. number of nodes, number of
edges, density, diameter)2. Node rankings (e.g. degree, betweenness, closeness centrality)3. Edge rankings (e.g. weight, betweenness centrality)4. Node rankings in pairs (e.g. degree vs. betweenness, plotted
on a scatter gram)5. Edge rankings in pairs6. Cohesive subgroups (e.g. finding communities in networks)7. Multiplexity (e.g. analyzing comparisons between different
edge types, such as friends vs. enemies)Shneiderman, Perer, Dunne
Micro ---- Scale ---- Macro
Micr
o ---
- Tim
e ---
- Mac
ro
Whole Graph
/ All Time
Sub Graph /
Narrow Time Slice
Sub Graph /
All Time
Whole Graph
/ Narrow Time
Slice
Filtering Network Diagrams
NodeXL Next Steps
• Enhanced layouts• Better manual layout of nodes• Clustering and composite nodes– Add/remove a node to/from a cluster– Add/remove a node to/from a composite
• Web interface (HTML5/Azure)• Add social network data sources:– Exchange, Sharepoint, Active Directory, ACM Digital
Library, , etc.
SOCIAL MEDIA NETWORK ANALYSIS WITH NODEXL
Network Overview, Discovery and Exploration for Excel 2007
Marc A. SmithChief Social ScientistConnected Action Consulting Group
[email protected]://www.ConnectedAction.nethttp://www.codeplex.com/nodexlhttp://www.twitter.com/marc_smithhttp://delicious.com/marc_smith/Paper http://www.flickr.com/photos/marc_smithhttp://www.facebook.com/marc.smith.sociologisthttp://www.linkedin.com/in/marcasmithhttp://www.slideshare.net/Marc_A_Smith