A presentation describing application of Node XL into analyzing social networks.
Made as part of project work for ITB course at VGSOM IIT Kharagpur.
By : Mayank Mohan
Anuradha Chakraborty
( Batch of 2012)
2. WHAT IS NODEXL?
“NodeXL is a free, open-source template for
Microsoft® Excel® 2007 and 2010 that makes it
easy to explore network graphs. With NodeXL, a
network edge list can be entered in a worksheet, a
button can be clicked and graph can be seen, all in
the familiar environment of the Excel window”
WHO WILL REQUIRE IT?
Students who are learning social media network
analysis
Professionals who are interested in applying
network analysis to business problems
3. WHY IS IT EASY-TO-USE AND APPROPRIATE?
Builds on the familiar spreadsheet paradigm
Provide an easy-to-use tool for nonprogrammers
A variety of visual properties
Supports powerful filtering
Calculates frequently used network metrics
Offers rich support for diverse visual network layouts
Includes powerful automated features, while allowing for
manual control graphical design
Integrates metrics, statistical methods, and visualization:
gains the benefit of all three approaches.
Supports work with modest-sized networks of several
thousand vertices
4. STEP 1: IMPORT DATA
To know how to import
data, watch our video:
http://youtu.be/39yXz7
2qdow
7. FACEBOOK
TWITTER
YOUTUBE
MAIL
FLICKR
and a lot of other social networking sites..
8. Search word:
“Kahaani”
Limit : 300
We effectively
found out the
two people
who were
most effective
in the
conversation
9. Search word: “Blood
Cancer”
Limit : 300
It is evident that
Social messages are
more re-tweeted
than advertisements
or other breaking
news
We effectively found
out the two people
who were most
effective in the
conversation
10. And finally…
We searched respectively
1. “Nike Discount”
2. “Adidas Discount”
Search Limit: 300
What have we found?
12. Search word:
“Kahaani”
Limit : 300
We effectively
found out the
two people
who were
most effective
in the
conversation
13. Search word:
“Kahaani”
Limit : 300
We effectively
found out the
two people
who were
most effective
in the
conversation
14. Search word:
“Kahaani”
Limit : 300
We effectively
found out the
two people
who were
most effective
in the
conversation
15. Search word:
“Kahaani”
Limit : 300
We effectively
found out the
two people
who were
most effective
in the
conversation
16. Search word:
“Kahaani”
Limit : 300
We effectively
found out the
two people
who were
most effective
in the
conversation
17. Search word:
“Kahaani”
Limit : 300
We effectively
found out the
two people
who were
most effective
in the
conversation
18.
19. Flexible Import and Export Import and export graphs in GraphML, Pajek, UCINet,
and matrix formats
Direct Connections to Social Networks Import social networks directly from Twitter,
YouTube, Flickr and email, or use one of several available plug-ins to get networks
from Facebook, Exchange and WWW hyperlinks.
Zoom and Scale Zoom into areas of interest, and scale the graph's vertices to reduce
clutter.
Flexible Layout Use one of several "force-directed" algorithms to lay out the graph,
or drag vertices around with the mouse. Have NodeXL move all of the graph's smaller
connected components to the bottom of the graph to focus on what's important.
Easily Adjusted Appearance Set the color, shape, size, label, and opacity of
individual vertices by filling in worksheet cells, or let NodeXL do it for you based on
vertex attributes such as degree, betweenness centrality or PageRank.
Dynamic Filtering Instantly hide vertices and edges using a set of sliders—hide all
vertices with degree less than five, for example.
Powerful Vertex Grouping Group the graph's vertices by common attributes, or have
NodeXL analyze their connectedness and automatically group them into clusters.
Make groups distinguishable using shapes and color, collapse them with a few clicks,
or put each group in its own box within the graph. "Bundle" intergroup edges to
make them more manageable.
Graph Metric Calculations Easily calculate degree, betweenness centrality, closeness
centrality, eigenvector centrality, PageRank, clustering coefficient, graph density and
more.
Task Automation Perform a set of repeated tasks with a single click
20. WHAT IS A NETWORK?
“A network or graph is defined as a collection
of n nodes connected by m edges.
A network can be
◦ Directed: The edges point in one direction
◦ Undirected: The edges go in both directions
Hypergraph :The graphs where the edges can
join more than two vertices together.
The edges can be weighted, contain self
loops, and have different properties within
the edges or nodes
21. VERTICES
In graph theory, a vertex (plural vertices) or node is the
fundamental unit out of which graphs are formed: an undirected
graph consists of a set of vertices and a set of edges (unordered
pairs of vertices), while a directed graph consists of a set of
vertices and a set of arcs (ordered pairs of vertices)
EDGES
A set of two vertices or an ordered pair, in the case of a directed
graph
An edge (a set of two elements) is drawn as a line connecting two
vertices, called endpoints or (less often) end-vertices
An edge with end-vertices x and y is denoted by xy (without any
symbol in between)
The edge set of G is usually denoted by E(G), or E when there is
no danger of confusion.
The size of a graph is the number of its edges, i.e. |E(G)|
22. TYPES OF NETWORK ANALYSIS METRICS
AGGREGATE NETWORK METRICS
A number of metrics describe entire networks
In some cases, a single network is broken into several disconnected pieces, called
components
Example: Network density can be used to systematically compare
communities, helping analysts decide which communities are highly
connected and which are sparse
VERTEX SPECIFIC NETWORK METRICS
Identifies individuals’ positions within a network
Paramount among these is the set of centrality measures, which describe how a
particular vertex can be said to be in the “middle” of a network. It emerges from the
concept that A person with fewer connections might have more “important”
connections than someone with more connections
The following centrality metrics provide quantifiable measures for these concepts:
◦ Degree Centrality
◦ Betweenness Centralities: Bridge Scores for Boundary Spanner
◦ Closeness Centrality: Distance Scores for Broadly Connected People
◦ Eigenvector Centrality: Influence Scores for Strategically Connected People
23. Degree Centrality
◦ Degree centrality is a simple count of the total number of
connections/edges linked to a vertex
◦ It can be thought of as a kind of popularity measure, but a crude one that
does not recognize a difference between quantity and quality
◦ For directed networks, there are two measures of degree: In-degree and
Out-degree
Betweenness Centralities: Bridge Scores for Boundary Spanner
◦ The distance between vertices who are not neighbors is measured by the
smallest number of neighbor-to-neighbor hops from one to the other
◦ Geo-desic Distance: the shortest path
◦ Example: a broker
Closeness Centrality: Distance Scores for Broadly Connected
People
◦ capturing the average distance between a vertex and every other vertex in
the network
Eigenvector Centrality: Influence Scores for Strategically
Connected People
◦ allows for connections to have a variable value, so that connecting to
some vertices has more benefit than connecting to other
◦ Example: Page Rank Algorithm by Google Search Engine
24. HOW ARE THE DATA REPRESENTED?
◦ Matrix
◦ Edge list
WHAT ARE THE TYPES OF NETWORK?
◦ Full and Partial Network
◦ Egocentric Network
◦ Unimodal Network
◦ Bimodal or Affiliation Network
◦ Multimodal Network
◦ Multiplex Network
25. HOW CAN NODEXL HELP USERS?
1. GRAPH METRICS:
a) Insights about a person’s position within the network,
helping to identify important or “central” people:
analysts and managers can better know who to contact
or influence or bring to the table when trying to
implement new programs or gain broader
understanding
b) identify cliques or persistent social roles that show up in
many communities
2. CLUSTERING
a) can help identify competing or complementary groups,
potential allies to form a powerful group, and
individuals who can connect you to a new group.
3. NATURE OF THE EGO
a) Social Media is a very low-cost advertisement medium
provided you know the track record of your ego