Introduction to Social Network Analytics

Social Network Analytics
(MIT-653)
-Asst. Prof. Bal Krishna Subedi, CDCSIT, TU
1

Chapter-1
1.1 Overview to Social Network Analytics
● Social Network Analytics (SNA) is an interdisciplinary field that combines
sociology, computer science, and data science to analyze and interpret the
relationships and interactions within networks.
● In the modern world, everything is connected in some way. Whether it's through
online social media, organizational structures, or communication systems,
relationships form the backbone of most systems.
● Networks provide a framework for understanding and analyzing the complex
relationships between different entities (often called nodes) through various types
of connections (called edges or ties).
● Understanding networks helps researchers and analysts to uncover hidden
patterns, predict behaviors, and make informed decisions across a range of
domains, including business, biology, sociology, economics, and communication
studies. 2

1.2 Importance of Social Network Analysis
1. Predicting behavior: By analyzing the relationships and interactions within a
network, it's possible to predict how certain behaviors might spread across a
population. This is valuable in areas like marketing, viral information
propagation, or disease transmission.
2. Optimizing resources: In organizational networks, understanding the flow of
information or tasks can improve efficiency and decision-making.
3. Enhancing communication: Analyzing the structure of communication
networks in a company or social media platform can help optimize channels
for faster, more effective communication.
3

1.3 What Are Networks?
1. A network is a collection of nodes (also called actors or vertices) that are connected by ties or
edges. Each node represents an entity, and each edge represents a relationship between
two nodes. The key aspect of networks is the relationship between the nodes. These
relationships can vary in type, strength, and direction.
There are two main types of networks:
● Social networks: Where the nodes represent individuals or organizations, and the edges
represent social relationships like friendship, trust, or communication.
● Technological networks: Where nodes are devices, and edges are communication
pathways like the internet or computer networks.
Networks can be represented mathematically as graphs, where the nodes are vertices, and
the edges (ties) are the connections between them.
4

1.4 Types of Networks:
1. Undirected Networks: The connections between nodes are bidirectional. If node
A is connected to node B, node B is also connected to node A. This is common in
social friendships, where friendship is mutual.
2. Directed Networks: Connections between nodes are one-way. For example, in a
Twitter network, if user A follows user B, it doesn’t necessarily mean that user B
follows user A.
3. Weighted Networks: The connections between nodes have weights, which can
represent the strength or frequency of interactions. For instance, the number of
calls between two people in a telecommunications network.
4. Bipartite Networks: A type of network in which nodes are divided into two disjoint
sets, with edges only between nodes of different sets. An example is a network of
authors and papers where each author is connected to the papers they wrote.
5

1.5 Example of a Social Network:
● In a social network like Facebook, each user (node) is connected to
others by friendships (edges). Analyzing this network can help
determine patterns of social influence, identify influential individuals
(central nodes), and even predict potential friendships or connections.
6

1.6 Types of Relations
1.6.1 Types of Relations in Social Networks:
● Friendship/Acquaintance: In social media networks, a tie might represent
a "friendship" or a "follow" relationship.
● Professional Relationships: In a corporate network, ties may represent
working relationships, such as collaborations or supervisory roles.
● Communication Networks: The ties could represent communication
between individuals, like emails, messages, or phone calls.
7

1.6.2 Types of Relations in Other Contexts:
● Biological Networks: In ecological or genetic networks, ties might
represent predator-prey relationships or gene interactions.
● Economic Networks: Relationships can represent trade or financial
transactions between businesses or countries.
8

1.6.4 Directed vs. Undirected Relations:
● Directed Relations: These are one-way relationships. For instance, in a
Twitter network, following someone is a directed tie because it’s not
necessarily reciprocal.
● Undirected Relations: These represent mutual relationships. An example
might be mutual friendships in Facebook.
9

1.6.4 Weighted and Unweighted Relations:
● Weighted Networks: Ties may have weights that quantify the strength or
frequency of the relationship. For example, the frequency of
communication between two individuals.
● Unweighted Networks: Ties either exist or don't, with no measure of
strength.
10

1.7 Goals of Social Network Analysis
Social Network analysis is conducted to answer various types of questions that help
researchers and analysts understand the structure and function of the network. Some key
goals of network analysis include:
1.7.1 Identifying Key Actors or Nodes:
Centrality measures: Central nodes (those with many connections) often hold
power or influence within the network. For instance, in a social media network,
influencers with many followers might be considered central actors.
11

1.7.2 Understanding Network Structure:
Clusters and Communities: One goal of network analysis is identifying
communities or subgroups within the network. These are groups of nodes that
are more densely connected to each other than to nodes outside the group.
12

1.7.3 Assessing Information Flow
Network Flow: In communication or transportation
networks, analyzing how information or goods flow through
the network can help optimize operations. For instance, in
social networks, we can track how information or memes
spread across the community.
13

1.7.4 Examining Network Dynamics
Evolution of Networks: Networks are often dynamic,
and studying their evolution over time can provide
insights into how connections form and disappear.
This is important in studying the growth of online
communities, the spread of diseases, or the formation
of new collaborations in scientific research.
14

1.7.5 Predicting Behavior
Network analysis can help predict individual or collective
behaviors, such as which individuals are likely to adopt
new products, who may be the next influencer, or how
rumors might spread through a community.
15

1.8 Network Variables as Explanatory Variables
In network analysis, the structure and dynamics of the network are studied to explain the
behavior or outcomes of individual nodes or the network as a whole. Network variables can
serve as explanatory variables in predictive or causal models.
1.8.1 Centrality as an Explanatory Variable
•Degree Centrality: The number of connections a node has. Nodes with high degree
centrality may be considered more important or influential. In the workplace, central
employees often have access to more information, which can influence their behavior and
decisions.
•Betweenness Centrality: Nodes that lie on the shortest path between many other nodes.
These nodes act as brokers or bridges, and their behavior might influence how information
or resources are distributed in the network.
•Closeness Centrality: Nodes that can reach all other nodes in the network in the shortest
number of steps. In communication networks, these nodes are likely to have quick access to
the entire network and can spread information efficiently.
16

In network analysis, the structure and dynamics of the network are studied to explain the
behavior or outcomes of individual nodes or the network as a whole. Network variables can
serve as explanatory variables in predictive or causal models.
1.8.1 Centrality as an Explanatory Variable
•Degree Centrality: The number of connections a node has. Nodes with high degree
centrality may be considered more important or influential. In the workplace, central
employees often have access to more information, which can influence their behavior and
decisions.
•Betweenness Centrality: Nodes that lie on the shortest path between many other nodes.
These nodes act as brokers or bridges, and their behavior might influence how information
or resources are distributed in the network.
•Closeness Centrality: Nodes that can reach all other nodes in the network in the shortest
number of steps. In communication networks, these nodes are likely to have quick access to
the entire network and can spread information efficiently.
17

1.8.2 Network Density
Network Density: The proportion of possible connections in a network that
actually exist. In a dense network, most nodes are interconnected, which
might influence the speed and nature of communication or collaboration.
18

1.8.3 Homophily
Homophily refers to the tendency of individuals to associate with others who are similar to
them. This concept can be used to explain the formation of communities or clusters within
social networks based on shared characteristics (e.g., interests, age, location).
19

1.9 Network Variables as Outcome Variables
In many cases, network analysis focuses on how the structure of a network influences
outcomes for individual nodes or the network as a whole. Here, network variables act as
outcome variables, helping us understand the consequences of network connectivity and
behavior.
1.9.1 Diffusion of Information
In a social network, a node’s centrality or the network’s density can affect how information
or influence spreads. For example, individuals in central positions may be more likely to
receive or disseminate information in a communication network.
1.9.2 Influence Propagation
Social networks can be used to predict how influence spreads from one person to another.
The structure of connections between individuals, especially those who are central, will
determine how influential certain individuals or ideas become.
20

1.9 Network Variables as Outcome Variables
1.9.3 Collective Action and Cooperation
Network structure can influence how likely individuals are to cooperate. For example, in a
network of employees, individuals who are more central might take the lead in
collaborative tasks or decisions, impacting the overall productivity and success of group
efforts.
1.9.4 Health Outcomes
In epidemiological networks, the outcome variable might be the spread of disease. The
structure of the network (e.g., how closely people are connected) can determine how
quickly a disease spreads across a population.
21

Summary:
Social networks are an essential framework for understanding the
complex interrelationships that drive behavior in various fields, from
social media to economics to biology. Understanding the why of
networks, the types of relations, the goals of analysis, and the roles that
network variables play as explanatory and outcome variables equips us
with the tools to model and interpret these complex systems.
Network analysis has far-reaching implications, from improving business
strategies to predicting the spread of diseases, and it is a valuable skill
for anyone interested in understanding the dynamics of connected
systems. Through the study of networks, we gain insight into the
underlying processes that shape behaviors, decisions, and outcomes
across different domains. 22

Introduction to Social Network Analytics

More Related Content

Similar to Introduction to Social Network Analytics (20)

More from Tribhuvan University (15)

Recently uploaded (20)

Introduction to Social Network Analytics