WOLFRAM NOTEBOOK

University of Illinois at Urbana - ChampaignDept . of Electrical and Computer Engineering​​

ECE 101 : Exploring Digital Information Technologies for Non - Engineers

Lecture 9: Social Networks

9.
1
A Social Media Poll

In[]:=
Out[]=

9.
2
Some Stats

In[]:=

9.
3
Some Stats

In[]:=

9.
4
What is Social Media?

◼
  • Online platforms that allow users to create, share, and interact with content and each other
    • ◼
  • Has become an integral part of our lives, influencing how we communicate, access information, and even perceive the world

    • 9.
    5
    Digital Mechanisms of Social Media

    ◼
  • User Profiles and Content Creation
    • ◼
  • Networking and Connectivity
    • ◼
  • Algorithms and Personalization
    • ◼
  • Privacy and Data Security

    • 9.
    6
    Social Networks on Social Media

    Graph Recap

    A graph is a collection of objects sharing some form of relationship. These objects are referred to as vertices (or nodes) and the relationships as edges (or links).

    Undirected, unweighted graph

    In[]:=
    Graph{sa,ad},
    Out[]=

    Directed, unweighted graph

    In[]:=
    Graph{s->a,a->d},
    Out[]=

    Undirected, weighted graph

    In[]:=
    Graph{sa,ad},EdgeWeight{30,10},
    Out[]=

    Directed, weighted graph

    In[]:=
    Graph{s->a,a->d},EdgeWeight{20,15},
    Out[]=

    Graph v.s Network

    ◼
  • Note: Not the network that is the Internet
    • ◼
  • Essentially the same. Graph used more in formal areas and Math. Network used more in applied areas.
    • ◼
  • A network usually refers to a graph that has been derived from a structured process and often has meaning in particular contexts.
    • ◼
  • A network can refer to a graph in which vertices and/or edges have properties.
    • ◼
  • A typical network might be an edge-weighted graph where the weights of edges refer to costs, lengths or capacities.

    • Examples of Networks

    Out[]=
    Computer sciences
    Database design, data mining, web mining, …
    Sociology
    Social networks, …
    Physics
    Models of phase transition, magnetism, …
    Ecology
    Food webs, …
    Economics
    Models of markets, sponsored search markets, …
    Business informatics
    Organisation diagrams, event-driven process chains, …
    Bioinformatics
    Models of epidemics, metabolic networks, …
    Graphs can be used to represent social networks
    ◼
  • Vertices are people
    • ◼
  • Edges are relationships
    • Graphs have properties that convey useful information.
    Number of vertices (number of people in this family):
    Number of edges (number of distinct relationships):
    List all the vertices (members of the family):
    List all the edges (the relationships):
    Number of edges associated with a particular vertex:
    Find the set of vertices with maximum vertex degree:
    The greatest distance between any pair of vertices:
    Shortest path between two vertices:
    Cliques are special graphs. Each vertex is connected to every other vertex.
    Human relationships—the social graph—form a graph now called a small-world graph, named after the expression, “It’s a small world!”
    Small world graphs have certain properties:
    ◼
  • small diameter and typical path length,
    • ◼
  • local cliques,
    • ◼
  • densely connected,
    • ◼
  • heavy tail

    • Small Diameter

    Any two people in the world are separated by short paths, on average about six steps.
    People who may be very far apart physically and socially are still connected with relatively small paths.
    This table is NOT meant to be precise, just to give you an idea of how slowly diameter grows.
    Most person-to-person distances are smaller.
    ◼
  • In late 2011, Facebook studied their network, which had around 720 million users at the time. Average shortest path length was 4.74 (Backstrom et al., 2011).

    • Is our family graph an example of a small world graph?

    Social networks inherit properties of small world graphs and capture human relationship
    ◼
  • local cliques (a person’s friends tend to know one another)
    • ◼
  • densely connected (most of the people have lots of connections)
    • ◼
  • heavy tails (some nodes have very high degree--the influencers--while other nodes have a reasonable number of connections)

    • Communities

    Communities in a graph are clumps of nodes that are more connected to each other than to the rest of the graph.
    Find a clique:
    Recreate the graph with the exact same connections as the original, but with the nodes arranged to illustrate the “community structure” of the graph:
    List the vertices:
    Most connected vertices:
    Vertices that are on many shortest paths of other vertex pairs--are important in maintaining the connectivity in the graph.

    Graph Mining

    A network of books linked by the same buyers on Amazon.com:
    Find sets of three books bought together:
    Find the largest selection of books frequently bought together that includes The Clinton Wars:
    Find the largest selection of books including The Clinton Wars that was frequently bought by buyers who previously bought a common book:

    Your turn ... Class Participation Question

    Answer the question asked in class
    How do companies store and provide access to a social graph?
    ◼
  • Algorithms analyze user behavior to curate personalized content feeds.
    • ◼
  • Collect information about your interactions and preferences
    • ◼
  • track various interactions such as likes, comments, shares, and time spent on posts
    • ◼
  • monitor how users engage with different types of content (e.g., videos, photos, articles) and which formats keep users engaged longer.
    • ◼
  • consider your search queries and the profiles or pages you visit.
    • ◼
  • data from your device and location to tailor content relevant to your geographical area
    • ◼
  • Personalization: Based on the collected data, algorithms rank content to prioritize what appears at the top of your feed.
    • ◼
  • Recommendations: Suggest new content, profiles, or pages to follow based on your interests and the behavior of similar users.
    • ◼
  • Targeted ads: Based on your interactions and preferences. E.g., if you frequently engage with fitness content, you might see ads for workout gear or health supplements.
    • ◼
  • Analyzing the Social Web by Jennifer Golbeck
    • ◼
    ​
    Wolfram Cloud

    You are using a browser not supported by the Wolfram Cloud

    Supported browsers include recent versions of Chrome, Edge, Firefox and Safari.


    I understand and wish to continue anyway »

    You are using a browser not supported by the Wolfram Cloud. Supported browsers include recent versions of Chrome, Edge, Firefox and Safari.