SlideShare a Scribd company logo

Small Worlds Social Graphs Social Media

Download as PPTX, PDF
••
suresh sood, profile picture
suresh sood

The document discusses what a small-world network is and provides context around the term. It references a 1967 psychology today article by Suresh Sood and Michael Light that may have originally coined the term. It also mentions that in 1991, the concept was popularized by studying social networks and the phenomenon of six degrees of separation. The small-world phenomenon shows that most people are connected through short chains of acquaintances.

1 of 66
Downloaded 13 times
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
What is a small-world network?Suresh Sood Michael Light (Miller Heiman)Psychology Today, vol. 1, no. 1, May 1967, pp61‐67?1991 Class of Harvard Law
GreatMystery14Suresh S.soodysoodywww.facebook.com/sureshsoodssoodHero5!twitter.com/soodywww.bravenewtalent.com/talent/suresh_soodwww.linkedin.com/in/sureshsoodGeektoidMangalascuzzy55suresh.sood@uts.edu.au
Giant Global Graph'll be thinking in the graph.My flights.My friends.Things in my life.My breakfast.What was that? Oh, yogourt, granola, nuts, and fresh fruit, since you ask.Submitted by timbl on Wed, 2007-11-21http://dig.csail.mit.edu/breadcrumbs/node/215
The BBC as Social GraphTopics EventsMusicProgrammesUsersGardeningNewsFood
Zuckerberg: “We Are Building A Web Where The Default Is Social”(3rd F8 developer conference, San Francisco)Open (Interest) Graph not only social connections between people, but connections people have with their interests—things, places, brands, and other sites. Yelp might create one around restaurants, Pandora might create one around music. Add some “like” buttons and anytime someone likes a restaurant or song anywhere on the Web with a Facebook like button, that information will flow back into the Open Graph. So that Yelp will know what restaurants you and your friends have liked elsewhere and take that into consideration when giving you recommendations, or Pandora with music, and so on. Facebook is taking some of the information that pops up in people’s realtime streams and baking it into the Web. “The stream is ephemeral,” says Zuckerberg. “It is there for a few hours and then it mostly floats away. Services don’t understand the semantic connections between you and that restaurant.” But now Facebook can. Instead of the Web being defined only by hyperlinks (to the benefit of search engines like Google), Facebook wants it to be defined by social connections, likes and dislikes, interests that are coded and machine-readable. “Our goal is to use the open graph so people can have instantly social experiences wherever they go,” he says.The Open Graph is hugely ambitious. Just wait until Facebook plugs in targeted advertising by: Location, Age, Sex, Keywords, Education, Workplace, Relationship Status, Relationship Interests & Languages.The Open Graph API will allow any page on the Web to have all the features of a Facebook Page. Once implemented, developers can include a number of Facebook Widgets, like the Fan Box, or enable the transformation of any Web page so it functions similar to a Facebook Page.
Small Worlds Social Graphs Social Media
Social Objects BloggingBusinessDatingPetsPhotosVideosReligiousSocial/Entertainment
Canada
Canadian Rockies
Tiffany Co.
LVMH – Louis Vuitton
Social Network RepresentationPrimary focus is actors & relationships # actors & attributesNodes (Actors) connected by Links (Ties/relationship or edge)Links represent flows or transfermaterial goods or information Adjacency list1: 22: 1, 33: 21Graph orsociogram23Adjacency matrix1 2 30 1 01 0 10 1 0123RelationshipActors1 = presence of link0 = no direct link
Key Network MeasuresDegree CentralityBetweenness CentralityCloseness CentralityEigenvector CentralityDiana’sCliquekrackkite.##h (modified labels)Connector(hub)VendorContractor ?BrokerBoundary spanners
NodeXL - Excel 2007 template for viewing and analyzing network graphswww.codeplex.com/NodeXL
UCINET 6 UCINET IV for DOS is free Grab bag of techniques and proceduresMatrix centered view rows & columns - actorscell value - relationshipCitation Borgatti, S.P., M.G. Everett, and L.C. Freeman. 1999. UCINET 6.0 Version 1.00. Natick: Analytic Technologies.Network analysis requires:##h file contains meta data about the network ##d file contains the actual data about the network
Useful ReferencesUCINET user guideTutorial Prof HannemanNetwork Analysis in Marketing (Webster & Morrison 2004)www.insna.org(international network for social analysis)
Data Language (DL) Filetypedl nr = 6, nc = 4 row labels embedded col labels embedded data: Dian Norm Coach Sam Mon 0 1 1 0 Tue 1 0 1 1 Wed 1 1 0 0 Thu 0 1 0 0 Fri 1 0 1 1  Sat 1 1 0 0
Standard Data SetsBERNARD & KILLWORTHFRATERNITY interactions among students living in a fraternity at a West Virginia collegeHAM RADIO radio calls made over a one-month period (voice-activated recording device)OFFICE interactions in a small business office. TECHNICALCAMP 92COUNTRIES TRADE DATADAVIS SOUTHERN CLUB WOMEN observed attendance at women’s club in 1930sFREEMAN'S EIES DATAGAGNON & MACRAE PRISONGALASKIEWICZ'S CEO'S AND CLUBSKAPFERER MINEKAPFERER TAILOR SHOPKNOKE BUREAUCRACIES 10 organizations and two relationships – money & info exchangeKRACKHARDT HIGH-TECH MANAGERSKRACKHARDT OFFICE CSSNEWCOMB FRATERNITYPADGETT FLORENTINE FAMILIESREAD HIGHLAND TRIBESROETHLISBERGER & DICKSON BANK WIRING ROOMSAMPSON MONASTERY Experimental and case study of social relationships." Doctoral dissertation, Cornell Univ.SCHWIMMER TARO EXCHANGESTOKMAN-ZIEGLER CORPORATE INTERLOCKSTHURMAN OFFICEWOLFE PRIMATESZACHARY KARATE CLUBBorgatti, S.P., Everett, M.G. and Freeman, L.C. 2002. Ucinet 6 for Windows. Harvard: Analytic Technologies.
Watts and Strogatz (1998)Collective Dynamics of Small-World Networks, Nature27 References represent history of the term “network” in systems and complexity sciencechemistrysystems biologyartificial neural networks Ecology systems sciencecellular automata dynamical systems theoryRobert Axelrod and Stuart Kauffman Pioneers of ComplexityStrogatz - dynamical systems theory (chaos)
20Modeling a social networkWatts – Strogatz (1998)Created a model for small-world networksLocal contactsLong-range contactsEffectively incorporated closed triads and short paths into the same model
Structured networkhigh clustering
large diameter
regularRandom networksmall clustering
small diameterSmall-world networkhigh clustering
small diameter
almost regularN = 1000 k=10D = 100 L = 49.51C = 0.67N =1000 k= 8-13D = 14 d = 11.1C = 0.63N =1000 k= 5-18D = 5 L = 4.46C = 0.01
LLCCNrandrand3.13.350.110.00023153127WWW3.652.990.790.00027225226Actors18.712.40.0800.0054914Power Grid2.652.250.280.05282C. ElegansDuncan J. Watts & Steven H. Strogatz,Nature393, 440-442 (1998)Real life networks are clustered, large C, but have small average distance L.
Duncan J. Watts & Steven H. Strogatz, Nature 393, 440-442 (1998)
Small Worlds Social Graphs Social Media
Duncan Watts - The “New” Science of Networks : Annu. Rev. Sociol. 2004. 30:243–70Research cluster builds on“…a long tradition of network analysis in sociology and anthropology (Degenne and Forse 1994; Scott 2000; Wasserman and Faust 1994)”“…an even longer history of graph theory in discrete mathematics (Ahuja et al. 1993; Bollobas 1998; West 1996)...” “spurred by the rapidly growing availability of cheap yet powerful computers and large-scale electronic datasets...”scholars come from a variety of disciplines, including “the mathematical, biological, and social sciences...”Watts notes the label “new science of networks” may “strike many sociologists as misleading, given the familiarity (to social network analysts) of many of its central ideas’’ (p. 342). ‘‘Nevertheless,’’ he argues, ‘‘the label does capture the sense of excitement surrounding what is unquestionably a fast developing field----new papers are appearing almost daily----and also the un-precedented degree of synthesis that this excitement has generated across the many disciplines in which network-related problems arise’’ (p. 243).
Duncan Watts - The “New” Science of Networks Annu. Rev. Sociol. 2004. 30:243–70Goal to make “substantial progress on a number of previously intractable problems, reformulating old ideas, introducing new techniques, and uncovering connections between what had seemed to be quite different problems” [Examples of intractable problems - analysis of large-scale, complex networks, the study of the evolution and transformation of complex networks over time, the study of how information, innovations, disease, cultural fads flow/move through complex networks. Old ideas - affiliation networks, the small-world problem and community structure. New techniques - discrete mathematics, the power law, cellular automata and agent-based modelling. Cross-disciplinary connections - similarities in network structure at different levels of scale, from a protein to a human organization to an ecosystem. ]
Traditional Practioners & New Networkers AnthropologistsSociologistsSocial workPsychologistsOrganizational theoristsPhysicistsMathematicians/StatisticiansEconomistsComputer ScientistsBiologistsEcologistsCriminologistsHealth Professionals
Key Scholars (New Science of Networks)Duncan Watts (new theoretic discovery: Small-World effect (Watts and Strogatz, 1998)Former Professor of sociology at Columbia University Principal research scientist at Yahoo! ResearchExternal faculty member of SFIAlbert-LászlóBarabási (new theoretic discovery Scale-Free feature (Barabási and Albert, 1999)Emil T. Hofman Professor of Physics and Director of the Center for Complex Networks at the University of Notre Dame, USA (http://www.nd.edu/~alb/)Associate of the Center of Cancer Systems Biology at the Dana Farber Cancer Institute, HarvardMark NewmanProfessor of Physics and Complex Systems at the University of MichiganExternal faculty member of the SFICenter for the Study of Complex Systems, University of Michigan ( John Holland /Robert Axelrod)Philip BonacichProfessor of Sociology at UCLAEditor of Journal of Mathematical SociologyBarry Wellman S.D. Clark Professor of Sociology, University of TorontoHead of NetLabInternational Coordinator, International Network for Social Network Analysis (INSNA)Creator creating INSNA (www.insna.org)
Barabási & Albert, Science286, 509 (1999)Scale-Free Model Barabasiand collaborators introduceScale-free networkswith two key rules: (a) growth in time by adding nodes and edges Starting with two connected nodes, add a new node to the network one at a time (b) preferential node attachment New nodes prefer to attach to the more connected nodes
Small Worlds Social Graphs Social Media
Small Worlds Social Graphs Social Media
April 2010 “Peer Influence Analysis”Two Types Of Mass Influencers
Small Worlds Social Graphs Social Media
Small Worlds Social Graphs Social Media
Susan Boyle“United Breaks Guitars”Old SpiceThe Man Your Man Could Smell Like Diasporahttp://www.nydailynews.com
Marketing in an Unpredictable WorldDuncan Watts & Steve Hasker, Harvard Business Review, September 2006“The implication for marketing executives is that they should de-emphasize designing, making, and selling would-be hits and focus instead on creating portfolios of products that can be marketed using real-time measurement of and rapid response to consumer feedback”.Target large number of ordinary individuals, and help them share messageTarget “big seed” of 10,000 people They recruit 5,000 extra people Those people recruit 2,500… Eventually dies out, but get 10,000 extra in processNetwork thinking replaces instinct The more we can measure and experiment , the more this will be true
Watts (2007), A twenty-first century science, Nature, 1 February, Vol 445, p.489“If handled appropriately, data about Internet-based communication and interactivity could revolutionize our understanding of collective human behaviour”.14,000 participants were asked to listen to, rate and download songs by unknown bands“Bored at work network” is millions of workers who share media, blog, and IM all dayFriend Sense Data (via Friend Sense app on Fb) 1500 respondents,17,500 complete dyads, 80,000 partials and 55,000 individual opinions traditional study cost USD200-300K and 2 years on Fb 2-3K and two months.
A Look at the NumbersWorldwide visitors to Twitter approached 10M in Feb 2009, up 700+% vs. Feb 2008 (Comscore)60+% stopped using Twitter a month after joining (Neilsen Online, via Reuters)Older than you think!18-24 year olds 12% less likely than average to visit Twitter25-54 year old crowd is driving this trend45-54 year olds 36% more likely to visit Twitter, making them the highest indexing age groupNext is 25-34 year olds: 30% more likelyTwitter rose to over 800,000 users in June 2009, up from 13,000 in 2008**2007 ~ 5,000 tweets per day2008, ~ 300,000 tweets per day2009 ~2.5 million tweets every dayEnd 2009 Tweet growth 1,400% reaching 35 million tweets per dayFeb 2010 Twitter sees 50 million tweets created per day or 600 TPSSource :Measuring Tweets, twitter blog, @kevinweil, viewed July 5 2010 <http://blog.twitter.com/2010/02/measuring-tweets.html>.Source: Reuters reporter Alexei Oreskovic.** comScore study, June 2009 Reported in Marketing Charts, August 17th 2009
AWashing Machine !
Week 2 statsTotal Researchers: 371 (+71 since last week)Total Active Data Collections: 262 (169 Streaming, 59 REST, 34 Curated Collections) (+37)Total finished Data Collections: 36 (+23)Total number of tweets in Database: ~33,234,149 (+15m)Total number of users in Database: ~14,437,195 (+7m)Number of tweets including the word “bieber”: too manyDatabase size: 28.7 GB (x2)Current rate of growth ~ 15 million tweets a week… or about 15 gigs a week.
140kit
Small Worlds Social Graphs Social Media
Social Media Marketing is not Conventional Marketing“a many-to-many mediated communications model in which consumers can interact with the medium, firms can provide content to the medium and, in the most radical departure from traditional marketing environments, consumers can provide commercially oriented content to the medium.”Hoffman & Novak, 1997
Small Worlds Social Graphs Social Media
Digital Breadcrumbs
Small Worlds Social Graphs Social Media
Small Worlds Social Graphs Social Media
Social Search It could represent a monumental shift in search technology. All major engines analyze the link structure of the Web as a key ingredient in determining what pages are most relevant -- a breakthrough that Google championed when it launched in 1998. A Web page that has a lot of other sites linking to it will rank higher, figuring more prominently in a given search, than one with only a few incoming links. Social search aims to shift power from Web publishers, who create these links, to everyday Internet users by examining their bookmarks or giving them tools to express their opinions.Current technology "delegates to Webmasters to decide what is important for the rest of us," says Bradley Horowitz, director of technology development at Yahoo. "Social search is about democratizing this power.”Yahoo Social Circle by Ben Elgin, January 23, 2006, Business Week
Facebook is My Newspaper(Susie Wilkening, http://reachadvisors.typepad.com/OneRiot.com: Search the realtime web
Small Worlds Social Graphs Social Media
Australia Leads Average Time Spent per Person on Social Media Sites in December 2009
Referral & Destination Traffic for April 2010agl.com.auSites people visit before going to agl.com.aufacebook.com (21.19%)google.com (15.57%)google.com.au ( 2.68%)Sites people visit after leaving agl.com.aufacebook.com (31.6%)energyaustralia.com.au (< 0.1%)google.com.au (<0.1%)
Wikipedia entries well placed on GoogleGenerate articles relating to your organisation,executives and newsMonitor articles on wikipedia for reputation managementReference with related entries
…Blogs are like conversations with friends. You share what you feel and what excites you about certain things. It's almost as good as being there. The fact that others can Google your topic and read is like tuning into a television station. We all want to know what's out there. Who's doing what, shopping where and what products help others. Blogs are just another way to share all the great things, not so great things and just a part of who we are. An outlet if you will. The blogisphere community is all connect and we make contacts in many ways. Through posts, through twitter conversations, through smaller nit community's, live web casts, and through conferences that we met in person. We make many friends and help each other with lot of topics. Many of us are Mom bloggers who stay at home and have no way of making new friends or communicating with others until we found blogging. Blogging creates friendships and that's what makes us real and connected.40 year old Mom blogger “nightowlmama” (#260)
55Linguistic Inquiry & Word Count : “junk words” in content http://www.liwc.net/liwcresearch07.phpCognitive complexity = zexcl + ztentat + znegate + zdiscrepzinclDepression = zI + zphyscal + znegemo – zposemoLiar = – zself – zother – zexcl + znegemo or Honesty = zself + zother + zexcl - znegemo - zmotionFemale = zself – zsixltr +z other + znegate – zarticle – zpreps + zcertain + zsocial + zpresent – zspace – zoccup + zhome – zmoneyAging = zposemo – zI + zsixltr + zcogmech + zexcl + zfuture – zpast – ztimePresidential = zsixltr – zwps – zunique – zpronoun – zself – zyou – zother – znegate + zarticle + zprepSlatcher, R.B., Chung, C.K., Pennebaker, J.W., & Stone, L.D. (2007), Winning words: Individual differences in linguistic style among U.S. presidential and vice presidential candidates, Journal of Research in Personality, 41, 63-75.
Linguistic Inquiry and Word Count (LIWC)Text Analysis : The Psychological Power of Words 56Pennebaker, J. W., Francis ME, Booth RJ. (2001). Linguistic Inquiry and Word Count (LIWC): LIWC2001. Mahwah: Lawrence Erlbaum Associates.
Brand Equity - Conversational Conversation Gap (Rubel 2005)Brand share of the online conversationGap between the total number of conversations about a category and the proportion which mention the brand operating in the categoryEquities of a Brand (Stein 2006)Topics being mentioned in conversations about a brand with equity share corresponding to the frequency at which each topic is mentionedSee pp 115-116 Cook, N 2008. Enterprise 2.0 Hampshire,England: Gower Publishing57
Conversation Gap - Vacation and Paris58* Total identified blogs: 99,181,005 @ 18 December, 2008
Paris – Equity Share Analysis of Attributes59* Total identified blogs: 99,181,005 @ 18 December, 2008
Blog Mentions Sydney Opera House, TaJMahal & Great Wall ChinaA review of the blogosphere on 8 June 2010 reveals 126.87 million blogs
8 Levels of Social Media Analyticshttp://manobyte.com/blog/index.php/2008/11/what-is-social-media-analytics orginally adapted from Davenport T (2007), Competing on Analytics

More Related Content

PDF
1999 ACM SIGCHI - Counting on Community in Cyberspace
Marc Smith
 
PDF
Mapping big data science
Han Woo PARK
 
PPTX
Meyer Big Data SDP13
Eric Meyer
 
PPT
Networking Theories Presentation
Leslie
 
PDF
빅데이터 시대의 미디어&커뮤니케이션 교육과 연구
Han Woo PARK
 
PDF
Mining the Social Web - Lecture 1 - T61.6020 lecture-01-slides
Michael Mathioudakis
 
PDF
What Actor-Network Theory (ANT) and digital methods can do for data journalis...
Liliana Bounegru
 
PPTX
04 Network Data Collection
Duke Network Analysis Center
 
1999 ACM SIGCHI - Counting on Community in Cyberspace
Marc Smith
 
Mapping big data science
Han Woo PARK
 
Meyer Big Data SDP13
Eric Meyer
 
Networking Theories Presentation
Leslie
 
빅데이터 시대의 미디어&커뮤니케이션 교육과 연구
Han Woo PARK
 
Mining the Social Web - Lecture 1 - T61.6020 lecture-01-slides
Michael Mathioudakis
 
What Actor-Network Theory (ANT) and digital methods can do for data journalis...
Liliana Bounegru
 
04 Network Data Collection
Duke Network Analysis Center
 

What's hot (20)

PPTX
2013 passbac-marc smith-node xl-sna-social media-formatted
Marc Smith
 
PDF
Ties
Guilherme Canteiro
 
PPT
Tools and Tips for Analyzing Social Media Data
Shelly D. Farnham, Ph.D.
 
PDF
Social network analysis intro part I
THomas Plotkowiak
 
PDF
Mining the Social Web - Lecture 2 - T61.6020
Michael Mathioudakis
 
PDF
Social Network Analysis & an Introduction to Tools
Patti Anklam
 
PPTX
Meyer dig ethno_2013sdp
Eric Meyer
 
PDF
10 Network Experiments
Duke Network Analysis Center
 
PPTX
Social Machines of Science and Scholarship
David De Roure
 
PPTX
Social Network Analysis Introduction including Data Structure Graph overview.
Doug Needham
 
PDF
Social network analysis & Big Data - Telecommunications and more
Wael Elrifai
 
PPT
The Basics of Social Network Analysis
Rory Sie
 
PPTX
Social Network Analysis: What It Is, Why We Should Care, and What We Can Lear...
Xiaohan Zeng
 
PPTX
20151001 charles university prague - marc smith - node xl-picturing political...
Marc Smith
 
PPTX
2013 Oxford Digital Humanities Summer School Workshop
Eric Meyer
 
PDF
Lecture 7: How to STUDY the Social Web? (2014)
Lora Aroyo
 
PDF
2014 ATHS Summer
Han Woo PARK
 
PPS
Social Technologies for Informaticians and Researchers
University of Michigan Taubman Health Sciences Library
 
PDF
Introduction to Social Network Analysis
Toronto Metropolitan University
 
PPT
Network theory
David Phillips
 
2013 passbac-marc smith-node xl-sna-social media-formatted
Marc Smith
 
Ties
Guilherme Canteiro
 
Tools and Tips for Analyzing Social Media Data
Shelly D. Farnham, Ph.D.
 
Social network analysis intro part I
THomas Plotkowiak
 
Mining the Social Web - Lecture 2 - T61.6020
Michael Mathioudakis
 
Social Network Analysis & an Introduction to Tools
Patti Anklam
 
Meyer dig ethno_2013sdp
Eric Meyer
 
10 Network Experiments
Duke Network Analysis Center
 
Social Machines of Science and Scholarship
David De Roure
 
Social Network Analysis Introduction including Data Structure Graph overview.
Doug Needham
 
Social network analysis & Big Data - Telecommunications and more
Wael Elrifai
 
The Basics of Social Network Analysis
Rory Sie
 
Social Network Analysis: What It Is, Why We Should Care, and What We Can Lear...
Xiaohan Zeng
 
20151001 charles university prague - marc smith - node xl-picturing political...
Marc Smith
 
2013 Oxford Digital Humanities Summer School Workshop
Eric Meyer
 
Lecture 7: How to STUDY the Social Web? (2014)
Lora Aroyo
 
2014 ATHS Summer
Han Woo PARK
 
Social Technologies for Informaticians and Researchers
University of Michigan Taubman Health Sciences Library
 
Introduction to Social Network Analysis
Toronto Metropolitan University
 
Network theory
David Phillips
 
Ad

Similar to Small Worlds Social Graphs Social Media (20)

PPTX
20120301 strata-marc smith-mapping social media networks with no coding using...
Marc Smith
 
PPTX
The Science Of Social Networks
Ehren Foss
 
PDF
Netwoks icml09
zhangzhao
 
PPTX
Module1:Social Networks-PG(Computer Network Engineering)
mayutapkire
 
PPT
System dynamics prof nagurney
Houw Liong The
 
PPTX
Social Network Analysis - Lecture 4 in Introduction to Computational Social S...
Lauri Eloranta
 
PDF
NetWorkShop: Boston Facilitators Roundtable
Patti Anklam
 
PPTX
LSS'11: Charting Collections Of Connections In Social Media
Local Social Summit
 
PPTX
20111103 con tech2011-marc smith
Marc Smith
 
PPT
An Introduction to Network Theory
Socialphysicist
 
PPTX
02 Introduction to Social Networks and Health: Key Concepts and Overview
Duke Network Analysis Center
 
PPTX
lec3_socialnetwork_part1.pptx
datapro2
 
PDF
Data mining based social network
Firas Husseini
 
PPTX
20121001 pawcon 2012-marc smith - mapping collections of connections in socia...
Marc Smith
 
PPTX
01 Network Data Collection (2017)
Duke Network Analysis Center
 
PDF
Chapter 3.pdf
Kongu Engineering College, Perundurai, Erode
 
PDF
Introduction to Computational Social Science
Premsankar Chakkingal
 
PPT
It’s a “small world” after all
quanmengli
 
PPTX
social.pptx
NISHASOMSCS113
 
PPTX
2013 NodeXL Social Media Network Analysis
Marc Smith
 
20120301 strata-marc smith-mapping social media networks with no coding using...
Marc Smith
 
The Science Of Social Networks
Ehren Foss
 
Netwoks icml09
zhangzhao
 
Module1:Social Networks-PG(Computer Network Engineering)
mayutapkire
 
System dynamics prof nagurney
Houw Liong The
 
Social Network Analysis - Lecture 4 in Introduction to Computational Social S...
Lauri Eloranta
 
NetWorkShop: Boston Facilitators Roundtable
Patti Anklam
 
LSS'11: Charting Collections Of Connections In Social Media
Local Social Summit
 
20111103 con tech2011-marc smith
Marc Smith
 
An Introduction to Network Theory
Socialphysicist
 
02 Introduction to Social Networks and Health: Key Concepts and Overview
Duke Network Analysis Center
 
lec3_socialnetwork_part1.pptx
datapro2
 
Data mining based social network
Firas Husseini
 
20121001 pawcon 2012-marc smith - mapping collections of connections in socia...
Marc Smith
 
01 Network Data Collection (2017)
Duke Network Analysis Center
 
Chapter 3.pdf
Kongu Engineering College, Perundurai, Erode
 
Introduction to Computational Social Science
Premsankar Chakkingal
 
It’s a “small world” after all
quanmengli
 
social.pptx
NISHASOMSCS113
 
2013 NodeXL Social Media Network Analysis
Marc Smith
 
Ad

More from suresh sood (20)

PPTX
Getting to the Edge of the Future - Tools & Trends of Foresight to Nowcasting
suresh sood
 
PPTX
Bigdata AI
suresh sood
 
PPTX
Bigdata ai
suresh sood
 
PPTX
Data Science Innovations
suresh sood
 
PPTX
Foresight conversation
suresh sood
 
PPTX
Data science Innovations January 2018
suresh sood
 
PPTX
future2020
suresh sood
 
PPTX
Data science innovations
suresh sood
 
PPTX
Data Science Innovations : Democratisation of Data and Data Science
suresh sood
 
PPTX
Swarm jobs
suresh sood
 
PPTX
Netnography online course part 1 of 3 17 november 2016
suresh sood
 
PPTX
Beyond dashboards
suresh sood
 
PPTX
Foresight Analytics
suresh sood
 
PPTX
Systemof insight
suresh sood
 
PPTX
TPA
suresh sood
 
PPT
Datapreneurs
suresh sood
 
PPT
Future of jobs, big data & innovation
suresh sood
 
PPT
Jobs Complexity
suresh sood
 
PPTX
Spark Social Media
suresh sood
 
PPTX
Bigdatacooltools
suresh sood
 
Getting to the Edge of the Future - Tools & Trends of Foresight to Nowcasting
suresh sood
 
Bigdata AI
suresh sood
 
Bigdata ai
suresh sood
 
Data Science Innovations
suresh sood
 
Foresight conversation
suresh sood
 
Data science Innovations January 2018
suresh sood
 
future2020
suresh sood
 
Data science innovations
suresh sood
 
Data Science Innovations : Democratisation of Data and Data Science
suresh sood
 
Swarm jobs
suresh sood
 
Netnography online course part 1 of 3 17 november 2016
suresh sood
 
Beyond dashboards
suresh sood
 
Foresight Analytics
suresh sood
 
Systemof insight
suresh sood
 
TPA
suresh sood
 
Datapreneurs
suresh sood
 
Future of jobs, big data & innovation
suresh sood
 
Jobs Complexity
suresh sood
 
Spark Social Media
suresh sood
 
Bigdatacooltools
suresh sood
 

Small Worlds Social Graphs Social Media

  • 1. What is a small-world network?Suresh Sood Michael Light (Miller Heiman)Psychology Today, vol. 1, no. 1, May 1967, pp61‐67?1991 Class of Harvard Law
  • 3. Giant Global Graph'll be thinking in the graph.My flights.My friends.Things in my life.My breakfast.What was that? Oh, yogourt, granola, nuts, and fresh fruit, since you ask.Submitted by timbl on Wed, 2007-11-21http://dig.csail.mit.edu/breadcrumbs/node/215
  • 4. The BBC as Social GraphTopics EventsMusicProgrammesUsersGardeningNewsFood
  • 5. Zuckerberg: “We Are Building A Web Where The Default Is Social”(3rd F8 developer conference, San Francisco)Open (Interest) Graph not only social connections between people, but connections people have with their interests—things, places, brands, and other sites. Yelp might create one around restaurants, Pandora might create one around music. Add some “like” buttons and anytime someone likes a restaurant or song anywhere on the Web with a Facebook like button, that information will flow back into the Open Graph. So that Yelp will know what restaurants you and your friends have liked elsewhere and take that into consideration when giving you recommendations, or Pandora with music, and so on. Facebook is taking some of the information that pops up in people’s realtime streams and baking it into the Web. “The stream is ephemeral,” says Zuckerberg. “It is there for a few hours and then it mostly floats away. Services don’t understand the semantic connections between you and that restaurant.” But now Facebook can. Instead of the Web being defined only by hyperlinks (to the benefit of search engines like Google), Facebook wants it to be defined by social connections, likes and dislikes, interests that are coded and machine-readable. “Our goal is to use the open graph so people can have instantly social experiences wherever they go,” he says.The Open Graph is hugely ambitious. Just wait until Facebook plugs in targeted advertising by: Location, Age, Sex, Keywords, Education, Workplace, Relationship Status, Relationship Interests & Languages.The Open Graph API will allow any page on the Web to have all the features of a Facebook Page. Once implemented, developers can include a number of Facebook Widgets, like the Fan Box, or enable the transformation of any Web page so it functions similar to a Facebook Page.
  • 11. LVMH – Louis Vuitton
  • 12. Social Network RepresentationPrimary focus is actors & relationships # actors & attributesNodes (Actors) connected by Links (Ties/relationship or edge)Links represent flows or transfermaterial goods or information Adjacency list1: 22: 1, 33: 21Graph orsociogram23Adjacency matrix1 2 30 1 01 0 10 1 0123RelationshipActors1 = presence of link0 = no direct link
  • 13. Key Network MeasuresDegree CentralityBetweenness CentralityCloseness CentralityEigenvector CentralityDiana’sCliquekrackkite.##h (modified labels)Connector(hub)VendorContractor ?BrokerBoundary spanners
  • 14. NodeXL - Excel 2007 template for viewing and analyzing network graphswww.codeplex.com/NodeXL
  • 15. UCINET 6 UCINET IV for DOS is free Grab bag of techniques and proceduresMatrix centered view rows & columns - actorscell value - relationshipCitation Borgatti, S.P., M.G. Everett, and L.C. Freeman. 1999. UCINET 6.0 Version 1.00. Natick: Analytic Technologies.Network analysis requires:##h file contains meta data about the network ##d file contains the actual data about the network
  • 16. Useful ReferencesUCINET user guideTutorial Prof HannemanNetwork Analysis in Marketing (Webster & Morrison 2004)www.insna.org(international network for social analysis)
  • 17. Data Language (DL) Filetypedl nr = 6, nc = 4 row labels embedded col labels embedded data: Dian Norm Coach Sam Mon 0 1 1 0 Tue 1 0 1 1 Wed 1 1 0 0 Thu 0 1 0 0 Fri 1 0 1 1  Sat 1 1 0 0
  • 18. Standard Data SetsBERNARD & KILLWORTHFRATERNITY interactions among students living in a fraternity at a West Virginia collegeHAM RADIO radio calls made over a one-month period (voice-activated recording device)OFFICE interactions in a small business office. TECHNICALCAMP 92COUNTRIES TRADE DATADAVIS SOUTHERN CLUB WOMEN observed attendance at women’s club in 1930sFREEMAN'S EIES DATAGAGNON & MACRAE PRISONGALASKIEWICZ'S CEO'S AND CLUBSKAPFERER MINEKAPFERER TAILOR SHOPKNOKE BUREAUCRACIES 10 organizations and two relationships – money & info exchangeKRACKHARDT HIGH-TECH MANAGERSKRACKHARDT OFFICE CSSNEWCOMB FRATERNITYPADGETT FLORENTINE FAMILIESREAD HIGHLAND TRIBESROETHLISBERGER & DICKSON BANK WIRING ROOMSAMPSON MONASTERY Experimental and case study of social relationships." Doctoral dissertation, Cornell Univ.SCHWIMMER TARO EXCHANGESTOKMAN-ZIEGLER CORPORATE INTERLOCKSTHURMAN OFFICEWOLFE PRIMATESZACHARY KARATE CLUBBorgatti, S.P., Everett, M.G. and Freeman, L.C. 2002. Ucinet 6 for Windows. Harvard: Analytic Technologies.
  • 19. Watts and Strogatz (1998)Collective Dynamics of Small-World Networks, Nature27 References represent history of the term “network” in systems and complexity sciencechemistrysystems biologyartificial neural networks Ecology systems sciencecellular automata dynamical systems theoryRobert Axelrod and Stuart Kauffman Pioneers of ComplexityStrogatz - dynamical systems theory (chaos)
  • 20. 20Modeling a social networkWatts – Strogatz (1998)Created a model for small-world networksLocal contactsLong-range contactsEffectively incorporated closed triads and short paths into the same model
  • 26. almost regularN = 1000 k=10D = 100 L = 49.51C = 0.67N =1000 k= 8-13D = 14 d = 11.1C = 0.63N =1000 k= 5-18D = 5 L = 4.46C = 0.01
  • 27. LLCCNrandrand3.13.350.110.00023153127WWW3.652.990.790.00027225226Actors18.712.40.0800.0054914Power Grid2.652.250.280.05282C. ElegansDuncan J. Watts & Steven H. Strogatz,Nature393, 440-442 (1998)Real life networks are clustered, large C, but have small average distance L.
  • 28. Duncan J. Watts & Steven H. Strogatz, Nature 393, 440-442 (1998)
  • 30. Duncan Watts - The “New” Science of Networks : Annu. Rev. Sociol. 2004. 30:243–70Research cluster builds on“…a long tradition of network analysis in sociology and anthropology (Degenne and Forse 1994; Scott 2000; Wasserman and Faust 1994)”“…an even longer history of graph theory in discrete mathematics (Ahuja et al. 1993; Bollobas 1998; West 1996)...” “spurred by the rapidly growing availability of cheap yet powerful computers and large-scale electronic datasets...”scholars come from a variety of disciplines, including “the mathematical, biological, and social sciences...”Watts notes the label “new science of networks” may “strike many sociologists as misleading, given the familiarity (to social network analysts) of many of its central ideas’’ (p. 342). ‘‘Nevertheless,’’ he argues, ‘‘the label does capture the sense of excitement surrounding what is unquestionably a fast developing field----new papers are appearing almost daily----and also the un-precedented degree of synthesis that this excitement has generated across the many disciplines in which network-related problems arise’’ (p. 243).
  • 31. Duncan Watts - The “New” Science of Networks Annu. Rev. Sociol. 2004. 30:243–70Goal to make “substantial progress on a number of previously intractable problems, reformulating old ideas, introducing new techniques, and uncovering connections between what had seemed to be quite different problems” [Examples of intractable problems - analysis of large-scale, complex networks, the study of the evolution and transformation of complex networks over time, the study of how information, innovations, disease, cultural fads flow/move through complex networks. Old ideas - affiliation networks, the small-world problem and community structure. New techniques - discrete mathematics, the power law, cellular automata and agent-based modelling. Cross-disciplinary connections - similarities in network structure at different levels of scale, from a protein to a human organization to an ecosystem. ]
  • 32. Traditional Practioners & New Networkers AnthropologistsSociologistsSocial workPsychologistsOrganizational theoristsPhysicistsMathematicians/StatisticiansEconomistsComputer ScientistsBiologistsEcologistsCriminologistsHealth Professionals
  • 33. Key Scholars (New Science of Networks)Duncan Watts (new theoretic discovery: Small-World effect (Watts and Strogatz, 1998)Former Professor of sociology at Columbia University Principal research scientist at Yahoo! ResearchExternal faculty member of SFIAlbert-LászlóBarabási (new theoretic discovery Scale-Free feature (Barabási and Albert, 1999)Emil T. Hofman Professor of Physics and Director of the Center for Complex Networks at the University of Notre Dame, USA (http://www.nd.edu/~alb/)Associate of the Center of Cancer Systems Biology at the Dana Farber Cancer Institute, HarvardMark NewmanProfessor of Physics and Complex Systems at the University of MichiganExternal faculty member of the SFICenter for the Study of Complex Systems, University of Michigan ( John Holland /Robert Axelrod)Philip BonacichProfessor of Sociology at UCLAEditor of Journal of Mathematical SociologyBarry Wellman S.D. Clark Professor of Sociology, University of TorontoHead of NetLabInternational Coordinator, International Network for Social Network Analysis (INSNA)Creator creating INSNA (www.insna.org)
  • 34. Barabási & Albert, Science286, 509 (1999)Scale-Free Model Barabasiand collaborators introduceScale-free networkswith two key rules: (a) growth in time by adding nodes and edges Starting with two connected nodes, add a new node to the network one at a time (b) preferential node attachment New nodes prefer to attach to the more connected nodes
  • 37. April 2010 “Peer Influence Analysis”Two Types Of Mass Influencers
  • 40. Susan Boyle“United Breaks Guitars”Old SpiceThe Man Your Man Could Smell Like Diasporahttp://www.nydailynews.com
  • 41. Marketing in an Unpredictable WorldDuncan Watts & Steve Hasker, Harvard Business Review, September 2006“The implication for marketing executives is that they should de-emphasize designing, making, and selling would-be hits and focus instead on creating portfolios of products that can be marketed using real-time measurement of and rapid response to consumer feedback”.Target large number of ordinary individuals, and help them share messageTarget “big seed” of 10,000 people They recruit 5,000 extra people Those people recruit 2,500… Eventually dies out, but get 10,000 extra in processNetwork thinking replaces instinct The more we can measure and experiment , the more this will be true
  • 42. Watts (2007), A twenty-first century science, Nature, 1 February, Vol 445, p.489“If handled appropriately, data about Internet-based communication and interactivity could revolutionize our understanding of collective human behaviour”.14,000 participants were asked to listen to, rate and download songs by unknown bands“Bored at work network” is millions of workers who share media, blog, and IM all dayFriend Sense Data (via Friend Sense app on Fb) 1500 respondents,17,500 complete dyads, 80,000 partials and 55,000 individual opinions traditional study cost USD200-300K and 2 years on Fb 2-3K and two months.
  • 43. A Look at the NumbersWorldwide visitors to Twitter approached 10M in Feb 2009, up 700+% vs. Feb 2008 (Comscore)60+% stopped using Twitter a month after joining (Neilsen Online, via Reuters)Older than you think!18-24 year olds 12% less likely than average to visit Twitter25-54 year old crowd is driving this trend45-54 year olds 36% more likely to visit Twitter, making them the highest indexing age groupNext is 25-34 year olds: 30% more likelyTwitter rose to over 800,000 users in June 2009, up from 13,000 in 2008**2007 ~ 5,000 tweets per day2008, ~ 300,000 tweets per day2009 ~2.5 million tweets every dayEnd 2009 Tweet growth 1,400% reaching 35 million tweets per dayFeb 2010 Twitter sees 50 million tweets created per day or 600 TPSSource :Measuring Tweets, twitter blog, @kevinweil, viewed July 5 2010 <http://blog.twitter.com/2010/02/measuring-tweets.html>.Source: Reuters reporter Alexei Oreskovic.** comScore study, June 2009 Reported in Marketing Charts, August 17th 2009
  • 45. Week 2 statsTotal Researchers: 371 (+71 since last week)Total Active Data Collections: 262 (169 Streaming, 59 REST, 34 Curated Collections) (+37)Total finished Data Collections: 36 (+23)Total number of tweets in Database: ~33,234,149 (+15m)Total number of users in Database: ~14,437,195 (+7m)Number of tweets including the word “bieber”: too manyDatabase size: 28.7 GB (x2)Current rate of growth ~ 15 million tweets a week… or about 15 gigs a week.
  • 48. Social Media Marketing is not Conventional Marketing“a many-to-many mediated communications model in which consumers can interact with the medium, firms can provide content to the medium and, in the most radical departure from traditional marketing environments, consumers can provide commercially oriented content to the medium.”Hoffman & Novak, 1997
  • 53. Social Search It could represent a monumental shift in search technology. All major engines analyze the link structure of the Web as a key ingredient in determining what pages are most relevant -- a breakthrough that Google championed when it launched in 1998. A Web page that has a lot of other sites linking to it will rank higher, figuring more prominently in a given search, than one with only a few incoming links. Social search aims to shift power from Web publishers, who create these links, to everyday Internet users by examining their bookmarks or giving them tools to express their opinions.Current technology "delegates to Webmasters to decide what is important for the rest of us," says Bradley Horowitz, director of technology development at Yahoo. "Social search is about democratizing this power.”Yahoo Social Circle by Ben Elgin, January 23, 2006, Business Week
  • 54. Facebook is My Newspaper(Susie Wilkening, http://reachadvisors.typepad.com/OneRiot.com: Search the realtime web
  • 56. Australia Leads Average Time Spent per Person on Social Media Sites in December 2009
  • 57. Referral & Destination Traffic for April 2010agl.com.auSites people visit before going to agl.com.aufacebook.com (21.19%)google.com (15.57%)google.com.au ( 2.68%)Sites people visit after leaving agl.com.aufacebook.com (31.6%)energyaustralia.com.au (< 0.1%)google.com.au (<0.1%)
  • 58. Wikipedia entries well placed on GoogleGenerate articles relating to your organisation,executives and newsMonitor articles on wikipedia for reputation managementReference with related entries
  • 59. …Blogs are like conversations with friends. You share what you feel and what excites you about certain things. It's almost as good as being there. The fact that others can Google your topic and read is like tuning into a television station. We all want to know what's out there. Who's doing what, shopping where and what products help others. Blogs are just another way to share all the great things, not so great things and just a part of who we are. An outlet if you will. The blogisphere community is all connect and we make contacts in many ways. Through posts, through twitter conversations, through smaller nit community's, live web casts, and through conferences that we met in person. We make many friends and help each other with lot of topics. Many of us are Mom bloggers who stay at home and have no way of making new friends or communicating with others until we found blogging. Blogging creates friendships and that's what makes us real and connected.40 year old Mom blogger “nightowlmama” (#260)
  • 60. 55Linguistic Inquiry & Word Count : “junk words” in content http://www.liwc.net/liwcresearch07.phpCognitive complexity = zexcl + ztentat + znegate + zdiscrepzinclDepression = zI + zphyscal + znegemo – zposemoLiar = – zself – zother – zexcl + znegemo or Honesty = zself + zother + zexcl - znegemo - zmotionFemale = zself – zsixltr +z other + znegate – zarticle – zpreps + zcertain + zsocial + zpresent – zspace – zoccup + zhome – zmoneyAging = zposemo – zI + zsixltr + zcogmech + zexcl + zfuture – zpast – ztimePresidential = zsixltr – zwps – zunique – zpronoun – zself – zyou – zother – znegate + zarticle + zprepSlatcher, R.B., Chung, C.K., Pennebaker, J.W., & Stone, L.D. (2007), Winning words: Individual differences in linguistic style among U.S. presidential and vice presidential candidates, Journal of Research in Personality, 41, 63-75.
  • 61. Linguistic Inquiry and Word Count (LIWC)Text Analysis : The Psychological Power of Words 56Pennebaker, J. W., Francis ME, Booth RJ. (2001). Linguistic Inquiry and Word Count (LIWC): LIWC2001. Mahwah: Lawrence Erlbaum Associates.
  • 62. Brand Equity - Conversational Conversation Gap (Rubel 2005)Brand share of the online conversationGap between the total number of conversations about a category and the proportion which mention the brand operating in the categoryEquities of a Brand (Stein 2006)Topics being mentioned in conversations about a brand with equity share corresponding to the frequency at which each topic is mentionedSee pp 115-116 Cook, N 2008. Enterprise 2.0 Hampshire,England: Gower Publishing57
  • 63. Conversation Gap - Vacation and Paris58* Total identified blogs: 99,181,005 @ 18 December, 2008
  • 64. Paris – Equity Share Analysis of Attributes59* Total identified blogs: 99,181,005 @ 18 December, 2008
  • 65. Blog Mentions Sydney Opera House, TaJMahal & Great Wall ChinaA review of the blogosphere on 8 June 2010 reveals 126.87 million blogs
  • 66. 8 Levels of Social Media Analyticshttp://manobyte.com/blog/index.php/2008/11/what-is-social-media-analytics orginally adapted from Davenport T (2007), Competing on Analytics
  • 67. YouTube Insight – Video Analytics
  • 68. A New Way of Marketing ? Social Media Marketing 1:1 Marketing ‘All Customersin a network interrelated’Segment Marketing‘All Customers are different’‘All Customers in a segment the same’ShotgunMarketing‘All Customers the same’
  • 69. New (Story Based) Segmentation The most powerful brand voice amongst consumersCrowdsourcing and Culture MappingApps = Segmentation Enablers e.g. Foursquare, FarmvilleHearts, Keys and Puppetry – Twitter Fairy TaleNeil Gaiman fantasy Writer124 Contributors over 8 days10,000 tweets  874 via editorial curationMarie Lenatupot and Tim Stock, What's next for segmentation? Admap Magazine, Feb 2010
  • 70. Levels of User EngagementCuratorsModerate a forumEdit a wiki
  • 71. Caution!“Children never put off till tomorrow what will keep them from going to bed tonight” ADVERTISING AGE

Editor's Notes

  • #14: Diana – max links (degree centrality) most connected – connector or hub – number of nodes connected – high influence of spreading info or virusHeather – best location powerful figure as broker to determine what flows and doesn’t –single point of failure – high betweeness = high influence – position of node as gatekeeper to exploit structural holes (gaps in network)Fernado &amp; Garth – shortest paths = closeness – the bigger the number the less centralEigenvector = importance of node in network ~ page rank google is similar measure
  • #49: The ability to disperse and share information through social platforms and do it using real-time tools is shifting the focus of content from “historical” news to real-time events.
  • #50: Also, think Mobile we will discuss UK stats shortly from December released over weekend. Social Media moving rapidly to be the gateway to web content Facebook has replaced her newspaper as the go-to place for relevant news in Susie’s life. It&apos;s not hard to imagine a near future where Facebook (and sites like it) also replace a lot of the ways we use atomized search.For people who are deeply immersed in social media, social networks are already a much heavier influence on personal choices--where to visit, what concert to attend--than traditional advertising. Which means that your organization&apos;s website--a brochure out in the wilderness of the Web--is only going to remain relevant and useful as a marketing piece if it is being referenced in the social context of your users&apos; lives. the next generation of Web users may find what they want by using their social network rather than a search algorithm. Social Media changes way we deal with web and in a sec we see mobile.
  • #54: http://wikifashion.com/wiki/Main_Page (see popular brands and new pages) – Brisbane based. &amp; wikiscannerHow Can You Use Wikipedia?•Create unbiased articles about your company, executives and important news stories&gt; Web site traffic increase•Cross-reference your articles with related articles &gt; New user discovery•Reputation managementSubscribe to updates to monitor changes Manage photos, links, content, etc.