ch6 (1)

Creating Value With Social Media Network Analytics

• Chapter 6 by Dr. Gohar F. Khan

• Copyright 2024 Gohar F. Khan

Learning Outcomes

• Gain knowledge and skills in:

  • Network Analytics Theories, Concepts, and Tools: Understanding how these can be applied effectively.

  • Types of Networks and Terminologies: Familiarity with varied network structures.

  • Business Intelligence Applications: Recognizing how social media analytics can support business strategies.

  • Conventional Social Media Network Analysis: Techniques for extraction, construction, and analysis of networks.

  • Network Strategies: Developing methods for effective networking.

Introduction

• Social media platforms create networks as users interact with each other and content

• These networks arise from components such as friendships, follows, content interactions, reviews, and professional connections, revealing valuable business insights.

What is Social Media Network Analytics?

• Definition: Involves constructing, analyzing, and comprehending social media networks.

• Purpose: Provides organizations with tools to decode networks' anatomy—size, structure, and key influencers.

Core Objectives of Network Analysis

• Overall Structure: Understanding the composition of networks, such as nodes and links.

• Influential Nodes: Identify and rank node significance based on connectivity.

• Pertinent Links: Assess relationships between nodes.

• Cohesive Subgroups: Discover tight-knit communities within larger networks.

• Multiplexity: Examine the interactions across different types of links.

Social Network Terminology

• Networks: Groups of interconnected nodes.

• Nodes: Represent entities such as individuals, organizations, or websites.

• Links: Illustrate the relationships among nodes, such as friendships or trade connections.

Types of Networks

• Social Networks:

  • Real World Example: Classmate network.

  • Online Example: Twitter following network.

• Social Network Sites: Software designed for creating and maintaining social relations; examples include Facebook and LinkedIn.

• Social Networking: The act of forming and maintaining social connections.

• Social Network Analysis: The scientific study of social networks, stemming from multiple fields like sociology and graph theory.

Network Structures

• Types of Structures:

  • Random Networks: Characterized by uniform node connectivity without dominant structures.

  • Scale-free Networks: Features power-law degree distributions where a few nodes have many connections (hubs).

  • Small World Networks: Consist of a few dominant nodes and many nodes with low connectivity, bridging gaps in networks.

Degree Distribution

• Definition: Probability distribution indicating the degree of connectivity of nodes in a network.

• Importance: Impacts information flow, node influence, and communication efficiency.

Random Networks

• Characteristics: Homogeneous degree distribution, low clustering coefficient, and short path lengths.

• Study Rationale: Benefits include benchmarking, theoretical insights, and understanding properties.

Scale-Free Networks

• Definition: Few central nodes dominate while many nodes possess fewer connections.

• Key Properties: Heterogeneous connectivity, robustness, and fragile structure under influence.

• Real-world Examples: Banking networks, Facebook, citation networks.

Small World Networks

• Characteristics: High clustering and short average path lengths; a few hubs dominate connectivity.

• Theory: Inspired by the 6-degree separation hypothesis, which states individuals are connected by a few links, enhanced through social media.

Social Media Network Topologies

• Different Forms of Network Structures: Emerge from user engagement on social media.

  • Types Include: Polarized Crowd, Tight Crowd, Brand Clusters, Broadcast Networks, Support Networks.

Common Social Media Network Types

• Co-authororship, Co-commenter, Geo Co-existence, Hyperlink, Friendship, Professional, and Dating Networks among others.

Network Types Based on Direction, Weight, and Mode

• Directed vs. Undirected: Placement of links determines if relationships have directionality.

• Weighted vs. Unweighted: Weights on links indicate strength of connections.

• One-Mode vs. Two-Mode Networks: Defined by the number of distinct node classes represented.

Node-Level Properties

• Degree Centrality: Measures connectivity based on direct links to other nodes.

• Betweenness Centrality: Identifies nodes that serve as vital bridges in the network.

• Eigenvector Centrality: Evaluates a node's influence based on the importance of its connections.

Network-Level Properties

• Clustering Coefficient: Indicates the tendency for nodes to form tightly connected groups.

• Density: Reflects the proportion of actual connections to possible connections, indicating network cohesion.

• Components: Subsets of nodes that are interconnected yet isolated from others.

• Diameter: Measures the longest shortest path between nodes, suggesting network spread.

Social Media Network Strategies

• Strategy 1: Focus on bridge building for enhanced connectivity.

• Strategy 2: Utilizing the shortest path to gain prominence within networks.

• Strategy 3: Crafting effective and engaging tweets.

Network Analytics Tools

• NodeXL: Visualization and analysis tool for social networks.

• UCINET: Robust application for network analysis with visualization capabilities.

• Pajek: Software for analyzing large networks.

• Netminer: Tool for social network analysis and visualization, particularly for education.

Review Questions

• Examples include defining types of networks and differentiating between network properties.

• Exercises: Focus on understanding network fundamentals and conducting network analysis.