Network Science

Bipartite graphs and Two-mode networks

“I think the next [21st] century will be the century of

 The simplest method of projecting a binary two-mode network is to select one set of
nodes and connect the nodes if they are connected with the same node of other set

 For example, in case of scientific collaboration network (Newman, 2001a) two scientists
are considered to be connected if they are the authors of the same paper from the set of
papers

 Through this method, one-mode network is created with entries representing the
presence of interconnection (ciβ =1 where at least one paper is shared) and c iβ =0 when there
is no sharing
2
 Types of network projection (Binary
method)
Binary method

 Through this method, one-mode network is created with entries representing the
presence of interconnection (ciβ =1 where at least one paper is shared) and c iβ =0 when there
is no sharing

3
 Types of network projection (Weighted
Binary method)
Weighted Binary method

 Further, we can transform two-mode binary network as weighted one-mode network by

For example, in case of scientific collaboration network, the number of papers those two
authors had collaborated on

Thus, the method is formalized as given in equation (Padro´n et al., 2011):

4
 Types of network projection (Weighted
Binary method)
Weighted Binary method

Where wij is the weight of link between node i and j, and p is the nodes of the other kind
that node i and node j are connected to, their co-occurrences

5
 Types of network projection (Newman
method)

 Newman argued (Newman, 2001b) and proposed another method during his research on
scientific collaboration network

He stressed on the point that the social bond weakens as the number of authors increase
in collaboration

According to him, collaboration can be formalized as given in equation

where, Np is the number of authors on paper p

6
 Types of network projection (Weighted
newman method)

 According to him, collaboration can be formalized as given in equation

where, Np is the number of authors on paper p

7
 Types of network projection (Weighted
newman method)
In the context of scientific collaboration networks, this implies that if two scientists who
only write a single paper together with no other co-authors get a weight of 1 (e.g., node B
and node D)

Moreover, if two scientists have written two papers together without any co-author, the
weight of their tie would be 2 (e.g., node B and E)

However, if the two scientists have a co-author, the weight on the tie between them is 0.5
(e.g., node A and node C)

A more complicated example is the tie between node A and node B in the diagram

8
 Types of network projection (Weighted
newman method)
 They have written two papers together: one without any other co-author and one with
node C as a co-author

The first paper would give their tie a weight of 1, and the second tie would add 0.5 to the
weight of this tie

Therefore, the weight attached to their tie is 1.5.

By discounting for the number of blue nodes attached to the same red node, this methods
creates a one-mode projection in which the strength of a node is equal to the number of ties
originating from that node in the two-mode network (e.g. the sum of weights attached to
ties originating from node A in the one-mode projection is 2, and node A is connected to
two red nodes)

9
Types of network projection (Sum method)
 Therefore, by including weight of links in the network make the analysis richer as
compared to network analysis with simple binary links in the network

 The weighted two-mode network projection can be formalized as given in equation

Where wi,p is the weight of ith node to pth event where i and j participated together

Through this method, it can be observed that how two nodes interact with common node
in different sets and project it onto a directed weighted one- mode network

This method will not create truly directed network as two directed links exists between
any two nodes connected with each other
10
Types of network projection (Sum method)

This method will not create truly directed network as two directed links exists between
any two nodes connected with each other

11
Types of network projection (Sum method)

 The Figure shows that user A has posted 4 messages in a blog where B has also posted 2
messages
 Similarly, 2 messages has been posted by user A in a blog where user B has posted 1
message
 Here, link number shows the weight as from user A to blog

 With such a method it is possible to differentiate how the two nodes interact with the
common node, and to project it onto a directed weighted one-mode network

 In this way, the data used for analysis is richer if the tie weights are included in the two-
mode network than if they are discarded

12
Types of network projection (Sum method)

 In this type of projected network, the weight of a link from a node to another node is not
necessarily the same as the weight attached to the link from the latter node to the former
node

It is not a truly directed network as two directed links exist between any connected node
pair

The direction shows the difference that how two nodes interact with the common node in
the network

We can say that 6 messages from (A) as compared to 3 messages from B as link weight

 Through this method, no any link is made up of single directed link (asymmetric) in
projected one-mode network 13
Types of network projection (Weighted
newman method)

 In a similar spirit as the method used by Newman (2001), it is also possible to discount
for the number of nodes when projecting weighted two-mode networks

For example, it could be argued that if many online users post to a thread, their ties
should be weaker than if there were few people posting to the thread

 The weights can be calculated based on this approach by using generalization as given in
equation

14
Types of network projection (Weighted
Newman method)

 This method would create a weighted directed one-mode network in which the out-
strength of a node is equal to the sum of the weights attached to the links in the two-mode
network that originated from that node

This will also not create truly directed network as two directed links exists between any
two nodes connected with each other

15
Types of network projection (Commands in
R)
# Load tnet
library(tnet)
 
# Binary one-mode projection
projecting_tm(net, method="binary")
 
# Simply the number of common red nodes
projecting_tm(net, method="sum")
 
# Newman's method
projecting_tm(net, method="Newman")

16