Topics In Social Computing (67810) Module 1 Introduction & The Structure of Social Networks.

Slides:



Advertisements
Similar presentations
Mobile Communication Networks Vahid Mirjalili Department of Mechanical Engineering Department of Biochemistry & Molecular Biology.
Advertisements

The Strength of Weak Ties
Milgram-Routing in Social Networks
Influence and Correlation in Social Networks Xufei wang Nov
Stelios Lelis UAegean, FME: Special Lecture Social Media & Social Networks (SM&SN)
Λ14 Διαδικτυακά Κοινωνικά Δίκτυα και Μέσα Strong and Weak Ties Chapter 3, from D. Easley and J. Kleinberg book.
Based on chapter 3 in Networks, Crowds and markets (by Easley and Kleinberg) Roy Mitz Supervised by: Prof. Ronitt Rubinfeld November 2014 Strong and weak.
Lecture 21 Network evolution Slides are modified from Jurij Leskovec, Jon Kleinberg and Christos Faloutsos.
Homophily, Social Influence, & Affiliation
Analysis of Social Media MLD , LTI William Cohen
Social Networks 101 P ROF. J ASON H ARTLINE AND P ROF. N ICOLE I MMORLICA.
Models of Network Formation Networked Life NETS 112 Fall 2013 Prof. Michael Kearns.
Nodes, Ties and Influence
Identity and search in social networks Presented by Pooja Deodhar Duncan J. Watts, Peter Sheridan Dodds and M. E. J. Newman.
CS 599: Social Media Analysis University of Southern California1 The Basics of Network Analysis Kristina Lerman University of Southern California.
Trends in Object-Oriented Software Evolution: Investigating Network Properties Alexander Chatzigeorgiou George Melas University of Macedonia Thessaloniki,
Directional triadic closure and edge deletion mechanism induce asymmetry in directed edge properties.
Link creation and profile alignment in the aNobii social network Luca Maria Aiello et al. Social Computing Feb 2014 Hyewon Lim.
Stress on Social Networks formed in a Mobile and Diverse World CS 460 December 8, 2005.
Networks and Organizations
Genetic Factors Predisposing to Homosexuality May Increase Mating Success in Heterosexuals Written by Zietsch et. al By Michael Berman and Lindsay Tooley.
Advanced Topics in Data Mining Special focus: Social Networks.
Graphs and Topology Yao Zhao. Background of Graph A graph is a pair G =(V,E) –Undirected graph and directed graph –Weighted graph and unweighted graph.
Reading Group “Networks, Crowds and Markets” Session 1: Graph Theory and Social Networks Typ hier de naam van de FEB afzender.
Friends and social networks. Announcements Treat Local and Cosmopolitan Networks Small Worlds (Milgram, Watts) Weak Ties (Granovetter)
A glimpse on social influence and link prediction in OSNs
Using Friendship Ties and Family Circles for Link Prediction Elena Zheleva, Lise Getoor, Jennifer Golbeck, Ugur Kuter (SNAKDD 2008)
Topic 13 Network Models Credits: C. Faloutsos and J. Leskovec Tutorial
The Best of Both Worlds of Psychology and Sociology
Λ14 Διαδικτυακά Κοινωνικά Δίκτυα και Μέσα Strong and Weak Ties Chapter 3, from D. Easley and J. Kleinberg book.
Λ14 Διαδικτυακά Κοινωνικά Δίκτυα και Μέσα Strong and Weak Ties Chapter 3, from D. Easley and J. Kleinberg book.
Analysis of Social Media MLD , LTI William Cohen
Data Analysis in YouTube. Introduction Social network + a video sharing media – Potential environment to propagate an influence. Friendship network and.
Homophily, Social Influence, & Affiliation Dr. Frank McCown Intro to Web Science Harding University This work is licensed under a Creative Commons Attribution-NonCommercial-
WALKING IN FACEBOOK: A CASE STUDY OF UNBIASED SAMPLING OF OSNS junction.
Principles of Social Network Analysis. Definition of Social Networks “A social network is a set of actors that may have relationships with one another”
Transitions... in your essay. Transition Words & Phrases Use transition words and phrases to show the direction of your thoughts. Use transition words.
1 The Spatial Dimension of Social Capital: An Exploration Zong-Rong Lee 李宗榮 Institute of Sociology Academia Sinica Taipei, Taiwan.
Online Social Networks and Media Homophilly Networks with Positive and Negative ties.
COLOR TEST COLOR TEST. Social Networks: Structure and Impact N ICOLE I MMORLICA, N ORTHWESTERN U.
Today’s topics Strength of Weak Ties Next Topic Acknowledgements
Using Graph Theory to Study Neural Networks (Watrous, Tandon, Conner, Pieters & Ekstrom, 2012)
Social Network Analysis Prof. Dr. Daning Hu Department of Informatics University of Zurich Mar 5th, 2013.
Λ14 Διαδικτυακά Κοινωνικά Δίκτυα και Μέσα Networks and Surrounding Contexts Chapter 4, from D. Easley and J. Kleinberg book.
A Computational Simulation to Study the Factors Impacting a College Student Body’s Racial Diversity Stephen Davies UMW Computer Science.
Susan O’Shea The Mitchell Centre for Social Network Analysis CCSR/Social Statistics, University of Manchester
Online Social Networks and Media
Comparing Snapshots of Networks Shah Jamal Alam and Ruth Meyer Centre for Policy Modelling 28 th March, 2007 – CAVES Bi-annual Meeting, IIASA,
What Is A Network? (and why do we care?). An Introduction to Network Theory | Kyle Findlay | SAMRA 2010 | 2 “A collection of objects (nodes) connected.
FANNIES DREAM What was she looking for? What was she like? What did she get? Why did she marry?
Does auto change residential segregation? Does telephone make us more or less sociable? Does TV viewing make us more violent? Does Internet use make us.
Internet Economics כלכלת האינטרנט Class 9 – social networks (based on chapter 3 from Easely & Kleinberg’s books) 1.
Selection & Social Influence Michael L. Nelson CS 495/595 Old Dominion University This work is licensed under a Creative Commons Attribution-NonCommercial-
Introduction Sample surveys involve chance error. Here we will study how to find the likely size of the chance error in a percentage, for simple random.
Social Networks Strong and Weak Ties
Topics In Social Computing (67810) Module 1 Introduction & The Structure of Social Networks.
Topics In Social Computing (67810) Module 1 Introduction & The Structure of Social Networks.
Graph clustering to detect network modules
Topics In Social Computing (67810)
The Strength of Weak Ties
Community detection in graphs
Models of Network Formation
Lecture 13 Network evolution
Models of Network Formation
Peer-to-Peer and Social Networks Fall 2017
Models of Network Formation
Models of Network Formation
Assortativity (people associate based on common attributes)
Social Network Analysis with Apache Spark and Neo4J
Lecture 21 Network evolution
Presentation transcript:

Topics In Social Computing (67810) Module 1 Introduction & The Structure of Social Networks

Class / Home experiment Who has a Facebook account? Pick a random Facebook “friend”. Is a Facebook “friend” really a friend? Check how many Facebook friends she/he has

Questions for thought How strong is your connection? Do you have many friends in common? Who has more friends (you or your friend)? – What influences this? – What would you expect the result to be for the “average” / person? – Are men and women different? Did you pick a friend from the same gender or a different one? Did you pick a friend with the same hair color or a different one?

STRUCTURE NETWORKS AS GRAPHS

Notation: Graphs

Undirected graphs uw v

Directed graphs uw v

Graphs have limitations that we will often ignore, but you shouldn’t forget: When do we place an edge? Strong / weak ties Undirected graphs: are relations really symmetric? Different types of connections Only represent relations between pairs. What about trios or more? Are both parties along an edge aware of the relationship it represents? Content ( s sent, content of webpages, other characteristics of edges/vertices)

Caltech Facebook network (2005) Still, full of data & hard to visualize

DO YOUR FRIENDS HAVE MORE FRIENDS THAN YOU? 11

Why your friends (probably) have more friends than you 12

Why your friends (probably) have more friends than you 13

A little more notation (Connectivity and cuts)

uw v

S Example: T

SOCIAL NETWORKS: CLUSTERED YET CONNECTED

Basic insight about social networks that we will explore from several angles: Social networks – are quite sparse – are highly “clustered” – tend to be “very well connected”

Triadic Closure Empirical evidence suggests: two friends (v,w) of a given individual (u) have an increased chance of being friends themselves. Many possible explanations: – Friends share same environment – Mutual friend introduces them – Attraction to similar people (Homophily) uw v

“Clustering and preferential attachment in Growing networks” M. E. J. Newman Scientific collaboration networks in Physics and Biology Probability of collaboration of scientists as a function of the number of previous mutual collaborators:

TouchGraph Visualization of Facebook friends and their friendships Notice the highly clustered structure…

Clustering coefficient

Example

Example

Clustering Coefficient

“The Anatomy of the Facebook Social Graph” Ugander et. al.

Birds of a feather flock together Homophily: The tendency of individuals to associate with “similar” individuals Studied since the 1920s in many contexts. Similarity: age, sex, ethnic background Links between similar individuals form more often, last longer and are more significant.

Examples: Similar age: Similar # of friends: From: “The Anatomy of the Facebook Social Graph” Ugander et. Al.

“Underlying” Similarity Must be careful to isolate the causes of homophily correctly Maybe people are friends with similar individuals simply because everyone in the area is similar? How do we compensate for this?

“Friendship in school: Gender and Racial Homophily”, Shrum et. al., Sociology of education Number of edges we’d expect to cross the cut

Racial segregation in high school / middle school social network

Mutable vs. Immutable Traits Socialization: choosing similar friends – (these may be similar due to both mutable & immutable traits) Social influence: Becoming more similar to friends (through acquired behavior) – E.g., taking up smoking because of friends (We will discuss the latter more in depth later in the course)

Spatial Effects (1) – Low Mobility In populations with low rates of geographic mobility relatives tend to live close to each other. Since relatives share genetic traits, this implies similarity between nearby individuals A strictly biological phenomena. – (niching, speciation, etc.)

Spatial Effects (2) – Schelling’s model of spatial segregation Simulated on an n-by-m grid. Each location is connected to its 8 neighbors. Place green / red agents at each grid point randomly, leaving some empty spaces

Call an agent “unhappy” if it sees only t=2 neighbors of its color or less. Pick an unhappy agent and move it to a nearby (or random) empty location. Repeat. Agents naturally form segregated clusters.

Spatial Effects (3) Distance & Friendships: The propinquity effect Friends are geographically close Attributed to increased probability of meeting Westgate community study – Housing for married MIT students returning from WW2 – Very homogenous From “Social Pressures in Informal Groups: A Study of Human Factors in Housing” By Leon Festinger, Kurt W. Back, Stanley Schachter

Distance#Friendships Max Possible #Friendships (at this dist.) ratio

Other physical properties that matter: – Route to apartment (stairs, doors) – Location of mail boxes – etc.

SOCIAL NETWORKS: CLUSTERED YET CONNECTED 39

Benefits of Connectivity Information obtained through links is valuable – If you are looking for a job it is beneficial to be connected to many potential sources – Many people find jobs through acquaintances. Tightly knit groups hold roughly the same information. Makes connections leading to other clusters highly valuable. 40

Mark Granovetter’s work on job search: – People find jobs through acquaintances Percentage of job finders Distance to job found 39.1%1 45.3%2 12.5%3 3.9%>3 41

Usually through weak links: Granovetter argues that weak links help more because they lead to distant parts of the social network. (But what’s missing in the table above?) Chance of getting a JobMeeting freq. with contact 16.7%Often (twice a week or more) 55.6%Occasionally (twice a week to once a year) 27.8%Rarely (once a year or less) 42

Bridges 43

Local Bridges A weaker property than a bridge: A local bridge: an edge that connects nodes that are otherwise of distance greater than 2. – A “shortcut” of sorts

Connection between weak links and triadic closure Granovetter’s idea: – strong links have triadic closure more often – Triadic closure adds links that reinforce network structure &“ruin” local bridges – Thus bridges are often weak links “The strength of weak links”

Connectivity as Social Capital Social capital: The ability to gain benefits through social relations. (Other use of the term: Social capital of a group rather than a person) Prediction: bridges confer an advantage. Hard to test! – Correlation between successful individuals and “bridging individuals” – But is this just the desire to connect to successful individuals? – Connections biased by homophily and selection processes 46

Several forms of (individual) social capital: Having many links – (incoming / outgoing links) – Having many links to successful individuals Short distance from many nodes in social network Belonging to / connecting many different communities 47