Social Media Analytics

Chapter 1. Social Media Analytics: An Overview

1. What are different types of social media?
Ans: In social media analytics, different types of social media are categorized based on their
function, content type, and interaction style. Here are the main types:
1. Social Networking Platforms: Connecting people and building relationships. Examples:
Facebook, LinkedIn, X (formerly Twitter)
2. Media Sharing Platforms: Sharing images, videos, and other media. Examples: Instagram,
YouTube, TikTok, Snapchat
3. Discussion Forums: Facilitating discussions and Q&A. Examples: Reddit, Quora, Stack
Overflow
4. Microblogging Platforms: Short-form updates, news, and opinions. Examples: X (Twitter),
Tumblr
5. Messaging Apps & Private Communities: Private and group communication. Examples:
WhatsApp, Telegram, Discord
6. Professional & Business Platforms: Business networking, hiring, and corporate branding.
Examples: LinkedIn, Glassdoor

2. What is social Media Landscape?

Ans: The social media landscape is diverse and enables people to connect and share in different
ways. People often use social media for different purposes. It includes various platforms like
Facebook, Instagram, Twitter, LinkedIn, TikTok, and others, each serving different purposes
such as news dissemination, professional networking, entertainment, and social interaction.
For example, they may visit one app to get the latest news, another to make professional
connections and another to discover what their favourite celebrities are doing.
This landscape constantly evolves with emerging trends, new technologies, and changing user
preferences. Understanding the social media landscape helps businesses, influencers, and
individuals effectively engage with their audiences and adapt to digital trends.

3. What is need of social media analytics? Explain in short SMA in small & large
organization.
Ans: Social Media Analytics (SMA) is essential for understanding audience behaviour,
measuring engagement, tracking trends, and making data-driven decisions. It helps
businesses optimize their strategies, improve customer interactions, and enhance brand
reputation.
Small Organizations:
  Helps in brand building with limited resources.
 Tracks audience engagement and campaign effectiveness.
 Identifies the best-performing content for better marketing strategies.
 Provides insights for cost-effective advertising and customer targeting.
Large Organizations:
 Analyses vast amounts of social data for trend prediction and brand reputation
management.
 Enhances customer service through sentiment analysis and feedback monitoring.
 Helps in competitor benchmarking and large-scale audience segmentation.
 Supports AI-driven insights for personalized marketing and business growth.

4. Differentiate social media vs. Traditional Business Analytics.

Ans:

5. Explain briefly seven layers of social media analytics.

Ans: The Seven Layers of Social Media Analytics refer to different dimensions of social
media data that help in analysing and understanding user behaviour, engagement, and trends.
the seven layers are:
1. Networks: Examines how users connect and interact across social platforms, including
relationships, communities, and influencer impact.
2. Text: Analyses the textual content of posts, comments, and messages to extract insights
such as sentiment, topics, and key themes.
 3. Actions: Focuses on user engagement metrics like likes, shares, comments, and
reactions to measure the impact of content.
4. Hyperlinks: Studies the links shared across social media to understand traffic sources,
content distribution, and referral patterns.
5. Location: Identifies the geographical distribution of users and social media
interactions, helping brands target specific regions effectively.
6. Search Engines: Analyses how social media content appears in search engine results
and its impact on online visibility and SEO.
7. Apps: Evaluates the role of third-party applications, mobile apps, and integrations in
social media engagement and user behaviour.
These seven layers provide a comprehensive approach to understanding social media data,
helping businesses and marketers optimize their strategies.

6. Draw social media Analytic Cycle & Explain in detail.

Ans: The Social Media Analytics Cycle is a structured process used to collect, analyse, and
interpret social media data to support business objectives. The social media Analytic Cycle
consists of six key stages:
1. Identification: The first step is to identify relevant social media data sources, such as
Facebook, Twitter, Instagram, LinkedIn, and YouTube. Businesses define their goals, such as
tracking brand mentions, customer feedback, or campaign performance.
2. Extraction: This stage involves gathering data from social media platforms using tools like
APIs, web scraping, or third-party analytics platforms. Data includes likes, shares, comments,
hashtags, and customer reviews.
3. Cleaning: Raw social media data often contains noise, such as spam, duplicate posts, and
irrelevant content. The data is filtered, organized, and formatted to improve accuracy and
relevance.
4. Analysing: This step involves processing and interpreting the cleaned data to generate
insights. Techniques like sentiment analysis, trend detection, and engagement metrics are used.
5. Visualization: Data is presented through charts, graphs, and dashboards for better
understanding and decision-making. Visualization helps stakeholders quickly interpret insights
and trends.
6. Interpretation: Businesses use the analysed and visualized data to derive actionable
insights. The results are used to improve marketing strategies, customer engagement, and brand
reputation.

7. What are the types of Social Media Analytics. What is predictive, Descriptive &
Prescriptive Analytics?
Ans: Social Media Analytics is broadly classified into four main types:
1. Descriptive Analytics – Answers "What happened?"
2. Diagnostic Analytics – Answers "Why did it happen?"
3. Predictive Analytics – Answers "What is likely to happen?"
4. Prescriptive Analytics – Answers "What should be done?"
1. Descriptive Analytics: Descriptive analytics focuses on summarizing past social media
performance using raw data. It helps in tracking key metrics like likes, shares, follower growth,
and engagement rates.
Example: A brand reviews its monthly social media report showing that it gained 5,000 new
followers and had an engagement rate of 10% on Instagram.
2. Diagnostic Analytics: Diagnostic analytics helps in understanding why certain trends or
outcomes occurred on social media by analysing patterns, behaviours, and correlations.
Example: A company sees a sudden drop in engagement and analyses that it was due to posting
at less active hours or sharing less relevant content.
3. Predictive Analytics: Predictive analytics uses historical data, machine learning, and AI to
forecast future trends and audience behaviours. It helps in proactive decision-making.
Example: Based on past data, an AI tool predicts that video content will receive 30% more
engagement than static images in the next marketing campaign.
4. Prescriptive Analytics: Prescriptive analytics provides actionable recommendations based
on predictive insights. It suggests the best strategies to improve social media performance.
Example: An analytics tool suggests that a brand should post at 6 PM on Wednesdays
because that is when their audience is most active.

8. Write short note on challenges of social media analytics.

Ans: Social media analytics provides valuable insights, but analyzing and interpreting the vast
amount of data comes with several challenges:
1. High Volume of Data: Social media generates massive amounts of data every second,
making it difficult to store, process, and analyse efficiently.
2. High Velocity: Data flows in real-time, requiring advanced tools and automation for
continuous monitoring and quick decision-making.
3. Data Diversity: Social media data includes text, images, videos, GIFs, and audio,
requiring advanced AI and machine learning techniques for proper analysis.
4. Unstructured Data Analysis: Unlike traditional business data, social media data is
largely unstructured, making it harder to categorize and interpret.
5. Sentiment Analysis Complexity: Understanding human emotions, sarcasm, and slang
in posts and comments is challenging for AI-driven sentiment analysis tools.
6. Privacy and Ethical Concerns: Collecting and analyzing social media data must
comply with data protection laws like GDPR and CCPA to ensure user privacy.
7. Lack of Advanced Tools & Skills: Many businesses lack the necessary AI-powered
tools and expertise required for real-time social media analytics.
8. Data Accuracy and Noise: Fake news, spam, and bot-generated content can distort
analytics results, making it difficult to extract meaningful insights.

9. List down social media analytics tools.

Ans:

Chapter 2. Social Network Structure, Measures & Visualization

1. What is degree distribution, density & connectivity in social network Structure?
Ans: a) Degree Distribution: Degree distribution refers to the number of connections (or
edges) each node (user or entity) has in a social network. It helps in understanding how
information spreads within the network.
Example: In Twitter, some users (influencers) have a high number of followers (high degree),
while most users have fewer followers (low degree).
b) Density: Density measures how interconnected the nodes in a network are. It is calculated
as the ratio of actual connections to the total possible connections. A higher density means more
interactions among users.
Example: A WhatsApp group of close friends has high density because most members interact
with each other frequently.
c) Connectivity: Connectivity determines how well different parts of a network are linked
together. A well-connected network allows for faster information flow.
Example: If a Facebook group has many subgroups that are well connected, information
spreads quickly across the entire community.

2. What is tie strength & Trust in SMA?

Ans: a) Tie Strength: Tie strength refers to the intensity of relationships between nodes in a
social network. Strong ties indicate close relationships, while weak ties connect distant
individuals.
Example: A Facebook friend you chat with daily has a strong tie. A LinkedIn connection
you’ve never interacted with has a weak tie.
b) Trust: Trust measures the reliability and credibility of nodes (users) in a social network. It
affects how people interact and share information.
Example: Verified accounts on Twitter are more trusted than anonymous accounts. Online
reviews from highly-rated users carry more weight than new accounts with no history.

3. Explain Centralization in social media analytics with example.

Ans: Centralization refers to the extent to which a social network is dominated by a few
influential nodes (users, accounts, or pages). A highly centralized network has a few dominant
nodes controlling most interactions, while a decentralized network has a more even distribution
of influence.
Example:
High Centralization: In Twitter, a few celebrities or influencers (like Elon Musk) dominate
discussions, and most users follow or interact with them.
Low Centralization: In Reddit communities, discussions are more decentralized, with many
users contributing equally to conversations.

4. What are scale issues in network visualization?

Ans: Network visualization is the graphical representation of relationships and interactions
within a network. It helps in understanding complex social structures, identifying key
influencers, and analyzing how information spreads.
Key Features of Network Visualization:
 Nodes (circles) represent individuals, brands, or entities.
 Edges (lines) represent connections, interactions, or relationships.
 Helps detect patterns, clusters, and central figures in a network.
Example:
 Twitter Network Visualization: Shows how users interact through mentions, retweets,
and replies.
 LinkedIn Network Graph: Displays professional connections and how they are
linked.
Scale issues occur when network visualizations become too large and complex, making them
difficult to interpret.
Common Scale Issues in Network Visualization:
1. Overlapping Nodes & Edges: When a network has too many connections, nodes and
edges overlap, making the visualization unreadable.
Solution: Use clustering techniques or filtering to simplify the graph.
2. Size & Complexity: Large-scale networks (millions of users) become difficult to
process and visualize.
Solution: Use aggregation methods to focus on key interactions.
 3. Scalability of Computation: Processing massive networks requires powerful
computing resources.
Solution: Use advanced graph databases and cloud-based visualization tools.
4. Data Interpretation Challenges: Large networks generate too much information,
making insights harder to extract.
Solution: Apply community detection algorithms to identify important subgroups.

5. List down Network Analytics Tools & Explain one in detail.

Ans: Here are some commonly used Network Analytics Tools for social media analysis:
1. NodeXL
2. UCINET
3. Pajek
4. Netminer
5. Flocker
6. Reach
7. Mentionmapp
NodeXL is an advanced network analysis and visualization tool designed for analyzing
relationships in social media platforms like Twitter, Facebook, and LinkedIn. It is widely used
by researchers, marketers, and businesses to study social connections and interactions.
Key Features of NodeXL:
 Graph Visualization: Provides an interactive graphical representation of social
networks.
 Data Import: Supports importing social media data directly from Twitter, YouTube,
and other platforms.
 Network Metrics Calculation: Measures degree distribution, centralization,
connectivity, and clustering.
 Sentiment Analysis: Analyzes tweets and social media posts to understand user
sentiment.
 Influencer Identification: Helps in detecting key influencers in a network.
Example Use Case: A company uses NodeXL to analyze Twitter conversations about its
brand. The tool helps identify top influencers, frequently used hashtags, and how information
spreads in the network.
 6.

Ans:
i) Is graph directed or undirected? How many nodes?
The graph is undirected because the edges do not have arrows indicating direction.
Number of nodes (vertices) = 8 (A, B, C, D, E, F, G, H).

ii) How many edges in graph?

By counting the connections (edges) between nodes:
(A-B), (B-C), (B-D), (C-E), (D-E), (E-F), (E-G), (E-H), (F-G), (G-H)
Total edges = 10

iii) Create an adjacency matrix for the graph.

The adjacency matrix is an 8×8 matrix, where:
1 indicates an edge exists between two nodes.
0 means no direct edge.
A B C D E F G H
A 0 1 0 0 0 0 0 0
B 1 0 1 1 0 0 0 0
C 0 1 0 0 1 0 0 0
D 0 1 0 0 1 0 0 0
E 0 0 1 1 0 1 1 1
F 0 0 0 0 1 0 1 0
G 0 0 0 0 1 1 0 1
H 0 0 0 0 1 0 1 0

iv) Estimate density of Graph.

 v) Create adjacency list for the Graph & what is largest clique in the network?
An adjacency list represents each node and its connected neighbours.
 A→B
 B → A, C, D
 C → B, E
 D → B, E
 E → C, D, F, G, H
 F → E, G
 G → E, F, H
 H → E, G

A clique is a subset of nodes where every node is connected to every other node.

 The largest clique in the network appears to be (E, G, H) since all three are fully
interconnected.

Thus, largest clique = {E, G, H} with 3 nodes.