Q.
1 Attempt the following questions (Any Five)
(A) Define Concept Mining and Sentiment Analysis.
1. Concept Mining:
○ Extracts concepts or themes from unstructured data.
○ Involves clustering similar terms and ideas.
○ Helps in understanding data patterns.
○ Useful for topic detection in social media.
○ Supports automatic categorization of content.
2. Sentiment Analysis:
○ Determines the sentiment (positive, negative, neutral) of text.
○ Uses NLP and machine learning for analysis.
○ Applied to social media for brand perception insights.
○ Useful in customer service and marketing.
○ Assists in tracking public opinion trends.
(B) Define Betweenness Centrality.
1. Measures the extent to which a node lies on paths between other nodes.
2. Indicates the node's influence in controlling information flow.
3. Higher betweenness centrality = greater influence in the network.
4. Calculated by the number of shortest paths passing through the node.
5. Helps in identifying critical nodes for communication or influence.
(C) Explain any two social media platforms.
1. Facebook:
○ Largest platform for personal networking and content sharing.
○ Supports text, image, video sharing, and live streaming.
○ Provides tools for groups, pages, events, and advertising.
○ Popular for community building and business engagement.
2. Twitter:
○ Microblogging platform with 280-character limit posts (tweets).
○ Known for real-time news and trend tracking.
○ Allows retweets, hashtags, and quick information dissemination.
○ Used widely for updates, customer interaction, and brand monitoring.
(D) Discuss some common social media network types.
1. Personal Networks (e.g., Facebook, LinkedIn): Connect with friends, family, and colleagues.
2. Content Sharing Networks (e.g., YouTube, Instagram): Share media like images and videos.
3. Discussion Forums (e.g., Reddit, Quora): Engage in discussions and ask questions.
4. Microblogging Platforms (e.g., Twitter): Share short posts, often in real-time.
5. Business Networks (e.g., LinkedIn): Focus on professional networking and job searching.
(E) Difference between Social Media Analytics & Traditional Business Analytics.
1. Data Type:
○ SMA: Unstructured, often text-based (e.g., tweets).
○ Traditional: Structured, numerical data (e.g., sales data).
2. Data Volume:
○ SMA: Large volume, high velocity (continuous social media posts).
○ Traditional: Comparatively lower volume and velocity.
3. Real-time Processing:
○ SMA: Often requires real-time analysis.
○ Traditional: Often batch processed over time.
4. Focus:
○ SMA: Customer sentiment, engagement, trends.
○ Traditional: Revenue, operational efficiency, productivity.
5. Tools:
○ SMA: NLP, machine learning.
○ Traditional: Statistical tools, business intelligence software.
(F) State the challenges of network visualization.
1. Scalability: Difficult to visualize large networks with many nodes.
2. Overlapping Nodes/Edges: Complex networks can appear cluttered.
3. Dynamic Nature: Social networks constantly change, requiring updates.
4. Node Labeling: Displaying labels for all nodes becomes challenging.
5. Interpretation: Making meaningful insights from a complex visual can be tough.
Q.2A Explain in detail about 1.5-degree ego-centric with an example.
1. Ego-Centric Network: Focuses on a single node (ego) and its direct and indirect connections.
2. 1-Degree Connection: Direct friends or followers of the ego node.
3. 1.5-Degree Connection: Direct connections plus friends of friends.
4. Visual Simplicity: Shows both immediate and slightly extended network.
5. Example: If "A" is the ego, "A" connects to B, C (1-degree), and B, C’s friends (1.5-degree).
Q.2B
(i) Draw a network graph showing your Facebook friends network with minimum 15 nodes.
(As this is an answer guide, drawing not provided here)
Answer the following questions with respect to your friend’s network graph.
1. Number of Nodes and Edges:
○ Nodes: Count the individuals.
○ Edges: Count the connections between individuals.
2. Graph Directed or Undirected:
○ Typically, Facebook friend networks are undirected (mutual friendship).
3. Density Calculation:
○
○ Provides insight into the network's connectivity.
4. Presence of Hubs:
○ Hubs are nodes with significantly higher connections.
○ Identify nodes with many links as they act as central points of connection.
Q.3A Enlist and explain different layout mechanisms for visualization of social networks.
1. Force-Directed Layout: Nodes repel each other while edges act like springs.
2. Circular Layout: Nodes arranged in a circle, showing relationships easily.
3. Hierarchical Layout: Nodes organized in levels (used for tree structures).
4. Grid Layout: Nodes aligned in a grid; useful for large networks.
5. Radial Layout: Nodes arranged radially, with ego at the center.
Q.3B Explain the STEPS of typical social media text analysis.
1. Data Collection: Gather social media posts, comments, or reviews.
2. Preprocessing: Clean data by removing stop words, special characters, etc.
3. Tokenization: Split text into individual words or tokens.
4. Sentiment Analysis: Detect the sentiment expressed (positive, negative, neutral).
5. Visualization and Interpretation: Present results through graphs and draw insights.
