SMA Pt1 Syllabus:

Chapter 1: Social Media Analytics: An Overview

●​ Core Characteristics of Social Media, Types of Social Media, Social
Media Landscape, Need for Social Media Analytics (SMA), SMA in small
& large organizations.
●​ Purpose of Social Media Analytics, Social Media vs. Traditional
Business Analytics, Seven Layers of Social Media Analytics, Types of
Social Media Analytics, Social Media Analytics Cycle, Challenges to
Social Media Analytics, Social Media Analytics Tools.

Chapter 2: Social Network Structure, Measures & Visualization

●​ Basics of Social Network Structure - Nodes, Edges & Tie.
 ●​ Describing the Networks Measures - Degree Distribution, Density,
Connectivity, Centralization, Tie Strength & Trust.
●​ Network Visualization - Graph Layout, Visualizing Network Features,
Scale Issues.
●​ Social Media Network Analytics - Common Network Terms, Common
Social Media Network Types, Types of Networks, Common Network
Terminologies, Network Analytics Tools.

Chapter 3: Social Media Text, Action & Hyperlink Analytics

●​ Social Media Text Analytics - Types of Social Media Text, Purpose of Text
Analytics, Steps in Text Analytics, Social Media Text Analysis Tools.
●​ Social Media Action Analytics - What Is Actions Analytics? Common
Social Media Actions, Actions Analytics Tools.
●​ Social Media Hyperlink Analytics - Types of Hyperlinks, Types of
Hyperlink Analytics, Hyperlink Analytics Tools.

Q1. Question Banks:

1.​ List the 5 people you are closest to on Facebook. Turn this list into a
network by listing all the connections between these people.
○​ Is your network directed or undirected?
○​ What do the edges represent (friendship, family relationship, close
relationships, acquaintances, etc.)?
○​ Give the adjacency list for the network.
○​ Give the adjacency matrix for the network.
○​ Are there any singletons?
2.​ Directed or Undirected Network:
○​ The network is undirected because Facebook friendships are mutual.
If Person A is friends with Person B, then Person B is also friends with
Person A.
3.​ What the Edges Represent:
○​ The edges represent friendship connections on Facebook.
4.​ Adjacency List:
○​ An adjacency list represents the network by listing each person and
their direct connections. For example:
i.​ Person A: Person B, Person C
ii.​ Person B: Person A, Person D
iii.​ Person C: Person A, Person E
iv.​ Person D: Person B
v.​ Person E: Person C
5.​ Adjacency Matrix:
 ○​ An adjacency matrix is a square matrix used to represent the network.
Rows and columns represent each person, and a 1 indicates a
connection, while a 0 indicates no connection. For example:
○​ Copy
ABCDE
A01100
B10010
C10001
D01000
E00100
6.​ Singletons:
○​ A singleton is a node with no connections. In this example, if any
person is not friends with any of the other four, they would be a
singleton. Based on the adjacency list and matrix provided, there are
no singletons.

1. Inlinks (if the network were directed):

●​ Definition: Inlinks are the edges that point toward a node. In a directed
network, an inlink to a node represents a connection where another node is
initiating the relationship.
●​ Example: If the network were directed, and Person B has an inlink from
Person A, it means Person A is "following" or "connecting to" Person B.
●​ In your case (undirected network):
○​ Since the network is undirected, there are no inlinks. All connections
are mutual, so the concept of inlinks does not apply.

2. Outlinks (if the network were directed):

●​ Definition: Outlinks are the edges that point away from a node. In a directed
network, an outlink from a node represents a connection where the node is
initiating the relationship.
●​ Example: If the network were directed, and Person A has an outlink to
Person B, it means Person A is "following" or "connecting to" Person B.
●​ In your case (undirected network):
○​ Since the network is undirected, there are no outlinks. All connections
are mutual, so the concept of outlinks does not apply.

3. Colinks (if the network were directed):

 ●​ Definition: Colinks refer to a situation where two nodes share a common
neighbor. In other words, two nodes are colinked if they both have a
connection to the same third node.
●​ Example: If Person A is connected to Person C, and Person B is also
connected to Person C, then Person A and Person B are colinked through
Person C.
●​ In your case (undirected network):
○​ Colinks can still be identified in an undirected network. For example:
■​ Person A and Person B are colinked through Person C (since
both are connected to Person C).
■​ Person B and Person D are colinked through Person A (since
both are connected to Person A).
■​ Person C and Person E are colinked through Person A (since
both are connected to Person A).

Numerical: Additional Questions:

Answer the following questions about this graph:

○​ a. How many nodes are in the network?
○​ b. How many edges are in the network?
○​ c. Is this graph directed or undirected?
○​ d. Create an adjacency list for this graph.
○​ e. Create an adjacency matrix for this graph.
○​ f. What is the length of the shortest path from node A to node F?
○​ g. What is the largest clique in this network? How many cliques of that
size are there?
 ○​ h. How many connected components are there in this network?
○​ i. Draw the 1.5 ego network for node E (without including node E in the
graph). How many singletons are in the ego network?
○​ j. Are there any hubs in the network? If so, which node(s) and why is it
a hub?

Q.2. Consider the hypothetical scenario within the social media

network LinkedIn, where Ramesh, an engineering graduate,
maintains connections with 10 of his friends. Ramesh is
represented as a node in the LinkedIn network, with each
connection reflecting a professional link to a friend. The structure
of this network, shaped by these connections, can be analyzed
using various network measures. Explain the following terms in the
context of the above case study:
○​ Degree distribution
○​ Density of connections
○​ Tie strength
○​ Hubs

https://docs.google.com/presentation/d/1lGrDuz4Iup_EQKCvDkRFV-auLsYP9p-9/edi
t#slide=id.p47

Q.3. (?)Why is it important for business managers to understand

and mine social media data?

Q.4. What is social media analytics, and how is it different from

traditional business analytics?
Q.5. Briefly explain the seven layers of social media data. Support
your answer with examples.
Q.6. Explain the social media analytics cycle. IECAVI
Q.7. (?) What ethical issues should be considered when mining
social media data?

Q.8. What are some main challenges to social media analytics?

1. Volume and Velocity:

●​ Volume: Social networks generate massive amounts of data due to the large
number of users and their interactions. This includes posts, comments, likes,
shares, and more.
●​ Velocity: The speed at which new data is generated and shared on social
networks is extremely high. Real-time analysis becomes challenging due to the
constant influx of data.

Impact: Handling such large-scale, fast-moving data requires robust computational

resources and efficient algorithms to process and analyze it in a timely manner.

2. Diversity Challenge:

●​ Multilingual Content: Social media users come from diverse linguistic

backgrounds, making it difficult to analyze content uniformly.
 ●​ Cultural and Contextual Variations: Content varies across different regions,
cultures, and time zones, adding complexity to understanding and interpreting
social interactions.
●​ Dynamic Nature: User behavior and content trends change over time, requiring
adaptive models for analysis.

Impact: Analysts must account for language translation, cultural context, and temporal
changes to ensure accurate and meaningful insights.

3. Unstructuredness as a Challenge:

●​ Unstructured Data: Social media data is often unstructured, including text,

images, videos, and emojis. This lack of structure makes it difficult to apply
traditional data analysis techniques.
●​ Noise and Irrelevance: Social media content often contains irrelevant or noisy
data (e.g., spam, advertisements), which can skew analysis results.

Impact: Advanced techniques like natural language processing (NLP), image recognition,
and machine learning are required to extract meaningful patterns from unstructured data.

Summary:

●​ Volume and Velocity: Managing large-scale, real-time data.

●​ Diversity: Addressing multilingual, multicultural, and dynamic content.
●​ Unstructuredness: Extracting insights from unstructured and noisy data.
 Q.9. Compare different social media analytics tools available in the
market and explain their strengths and weakness.

Summary of Strengths and Weaknesses:

Tool Strengths Weaknesses

DiscoverText Text analysis, data cleaning, Limited visualization,

qualitative research requires technical expertise
Lexalytics NLP, sentiment analysis, Expensive, text-focused
multilingual support

Netlytic Network analysis, real-time data, Limited advanced analytics,

user-friendly Twitter-focused

NodeXL Network visualization, Excel Limited scalability, requires

integration, multi-platform support Excel

Google Comprehensive web analytics, free, Limited social media

Analytics tracks user behavior insights

Mentionmapp Real-time monitoring, trending Twitter-focused, lacks

topics, user-friendly advanced analytics

Tweepsmap Location-based analytics, audience Twitter-focused, limited to

demographics geographic analysis

Google Tracks search trends, free, global Limited to search data, not
Trends insights social media-specific

UCINET Advanced network analysis, Steep learning curve,

academic research expensive

Mixpanel User behavior tracking, real-time Expensive, limited social

analytics, mobile/web focus media features

Numerical: 2. Consider this graph

1.​ What is the largest connected component?
2.​ Are there any bridges in the network? If so, where are they?
3.​ Create an adjacency list for the network.
4.​ Create an adjacency matrix for the network.
Q.12. What is text analytics, and why is it useful? Explain with an
example.
Q.13. Differentiate between static and dynamic social media text.
Explain with an example.

Difference Between Static and Dynamic Social Media Text

Feature Static Social Media Text Dynamic Social Media Text

Definition Text that remains unchanged Text that evolves, updates, or changes
over time. based on user interactions.

Nature Permanent and fixed once Frequently updated, edited, or interactive.

posted.
Examples Blog posts, Wikipedia entries, Live tweets, Facebook/Instagram stories,
archived tweets, news articles. comment sections, real-time discussions.

Interactio Limited or no changes after Changes based on likes, shares, replies, or

n posting. real-time data.

Purpose Provides static information for Engages users with continuous updates
reference. and responses.

Example

●​ Static Text: A Wikipedia article about a historical event remains unchanged unless
updated manually.
●​ Dynamic Text: A Twitter (X) thread about an ongoing sports event updates in
real-time with new scores, player stats, and audience reactions.

Q.14. Explain the four main purposes of social media text analytics.
Q.15. Explain the typical social media text analysis steps.

Q.16. (?)Define social media actions analytics.

Q.17. (?)Briefly list and define different actions performed by social

media users considering various types of social media platforms.

Q.18. (?)Why is it important to measure actions performed by social

media users?
Q.19. What are hyperlinks, and why are they important?

Q.20. Briefly discuss in-links, out-links, and co-links.

Q.21. What is hyperlink analytics and its underlying assumptions?
Q.22. What is hyperlink environment analysis?
Q.23. What is link impact analysis?
–Assumptions??