Social Network Analysis: What It Is, Why We Should Care, and What We Can Lear...Xiaohan Zeng
ย
The advent of the social networks has completely changed our daily life. The deluge of data collected on Social Network Services (SNS) and recent developments in complex network theory have enabled many marvelous predictive analysis, which tells us many amazing stories.
Why do we often feel that "the world is so small?" Is the six-degree separation purely imagination or based on mathematical insights? Why are there just a few rockstars who enjoy extreme popularity while most of us stay unknown to the world? When science meets coffee shop knowledge, things are bound to be intriguing.
I will first briefly describe what social networks are, in the mathematical sense. Then I will introduce some ways to extract characteristics of networks, and how these analyses can explain many anecdotes in our life. Finally, I'll show an example of what we can learn from social network analysis, based on data from Groupon.
Social Network Analysis Workshop
This talk will be a workshop featuring an overview of basic theory and methods for social network analysis and an introduction to igraph. The first half of the talk will be a discussion of the concepts and the second half will feature code examples and demonstrations.
Igraph is a package in R, Python, and C++ that supports social network analysis and network data visualization.
Ian McCulloh holds joint appointments as a Parsonโs Fellow in the Bloomberg School of Public health, a Senior Lecturer in the Whiting School of Engineering and a senior scientist at the Applied Physics Lab, at Johns Hopkins University. His current research is focused on strategic influence in online networks. His most recent papers have been focused on the neuroscience of persuasion and measuring influence in online social media firestorms. He is the author of โSocial Network Analysis with Applicationsโ (Wiley: 2013), โNetworks Over Timeโ (Oxford: forthcoming) and has published 48 peer-reviewed papers, primarily in the area of social network analysis. His current applied work is focused on educating soldiers and marines in advanced methods for open source research and data science leadership.
More information about Dr. Ian McCulloh's work can be found at https://ep.jhu.edu/about-us/faculty-directory/1511-ian-mcculloh
Social Network Analysis: What It Is, Why We Should Care, and What We Can Lear...Xiaohan Zeng
ย
The advent of the social networks has completely changed our daily life. The deluge of data collected on Social Network Services (SNS) and recent developments in complex network theory have enabled many marvelous predictive analysis, which tells us many amazing stories.
Why do we often feel that "the world is so small?" Is the six-degree separation purely imagination or based on mathematical insights? Why are there just a few rockstars who enjoy extreme popularity while most of us stay unknown to the world? When science meets coffee shop knowledge, things are bound to be intriguing.
I will first briefly describe what social networks are, in the mathematical sense. Then I will introduce some ways to extract characteristics of networks, and how these analyses can explain many anecdotes in our life. Finally, I'll show an example of what we can learn from social network analysis, based on data from Groupon.
Social Network Analysis Workshop
This talk will be a workshop featuring an overview of basic theory and methods for social network analysis and an introduction to igraph. The first half of the talk will be a discussion of the concepts and the second half will feature code examples and demonstrations.
Igraph is a package in R, Python, and C++ that supports social network analysis and network data visualization.
Ian McCulloh holds joint appointments as a Parsonโs Fellow in the Bloomberg School of Public health, a Senior Lecturer in the Whiting School of Engineering and a senior scientist at the Applied Physics Lab, at Johns Hopkins University. His current research is focused on strategic influence in online networks. His most recent papers have been focused on the neuroscience of persuasion and measuring influence in online social media firestorms. He is the author of โSocial Network Analysis with Applicationsโ (Wiley: 2013), โNetworks Over Timeโ (Oxford: forthcoming) and has published 48 peer-reviewed papers, primarily in the area of social network analysis. His current applied work is focused on educating soldiers and marines in advanced methods for open source research and data science leadership.
More information about Dr. Ian McCulloh's work can be found at https://ep.jhu.edu/about-us/faculty-directory/1511-ian-mcculloh
Clustering Methods and Community Detection with NetworkX. A slide deck for the NTU Complexity Science Winter School.
For the accompanying iPython Notebook, visit: http://github.com/eflegara/NetStruc
An introduction in the world of Social Network Analysis and a view on how this may help learning networks. History, data collection and several analysis techniques are shown.
Social Network Analysis power point presentation Ratnesh Shah
ย
Basics of social network analysis,Application and also explain interesting study done by facebook , twitter, youtube and many more social media network ,slide contains some of interesting study to get knowledge about online social network analysis.
What power law and rich get richer phenomena means in the world of network and how does it affect in the social networks for web page popularity especially in the facebook platform?
Clustering Methods and Community Detection with NetworkX. A slide deck for the NTU Complexity Science Winter School.
For the accompanying iPython Notebook, visit: http://github.com/eflegara/NetStruc
An introduction in the world of Social Network Analysis and a view on how this may help learning networks. History, data collection and several analysis techniques are shown.
Social Network Analysis power point presentation Ratnesh Shah
ย
Basics of social network analysis,Application and also explain interesting study done by facebook , twitter, youtube and many more social media network ,slide contains some of interesting study to get knowledge about online social network analysis.
What power law and rich get richer phenomena means in the world of network and how does it affect in the social networks for web page popularity especially in the facebook platform?
eBay Pulsar: Real-time analytics platformKyoungMo Yang
ย
http://blog.embian.com/74
Pulsar โ an open-source, real-time analytics platform and stream processing framework. Pulsar can be used to collect and process user and business events in real time, providing key insights and enabling systems to react to user activities within seconds. In addition to real-time sessionization and multi-dimensional metrics aggregation over time windows, Pulsar uses a SQL-like event processing language to offer custom stream creation through data enrichment, mutation, and filtering. Pulsar scales to a million events per second with high availability. It can be easily integrated with metrics stores like Cassandra and Druid.
Another Interdisciplinary Transformation: Beyond an Area-studies JournalHan Woo PARK
ย
Journal of Contemporary Eastern Asia (ISSN 2383-9449) is a refereed biannual journal that takes a lead on a new scholarship in Asia. In the past, the JCEA was dedicated to the study of current political, social and economic trends in East and Southeast Asia. But now, the JCEA finds unique aspects of Asian scholarship by expanding its scope to (socio-technical) convergence and future (network) studies. The JCEA editors are working very hard to boost the scholarly presence of new Asian scholarship around the world and secure its reputation as an emerging world-class publishing outlet. The editors welcome manuscripts based on original research or significant reexamination of existing literature.
Global mapping of artificial intelligence in Google and Google ScholarHan Woo PARK
ย
Omar, M., Mehmood, A., Choi, G.S., Park, H.W.@ (2017 Online First). Global mapping of artificial intelligence in Google and Google Scholar. Scientometrics.
https://link.springer.com/article/10.1007%2Fs11192-017-2534-4
#์ธ๊ณต์ง๋ฅย ๋ถ์ผ์ ๊ตญ๊ฐ๋ณ ํํฉ๊ณผ ์ ๋ง์ย #์น๋ณด๋ฉํธ๋ฆญ์คย #๋น ๋ฐ์ดํฐย #๊ตฌ๊ธ๋ฐ์ดํฐย ์ด์ฉํด์ ์ธ๊ณ์ง๋๋ฅผ ๊ทธ๋ ธ์ต๋๋ค.
๋ฐํ์ฐ ์์ด ์ด๋ ฅ์ Curriculum vitae ๊ฒฝํฌ๋ ํ์ฌ ์ ์ถ์ฉHan Woo PARK
ย
Full Prof. Dr. Han Woo PARK, http://www.hanpark.net https://www.slideshare.net/hanpark
(BA-HUFS, MA-Seoul Nat'l Univ. PhD-SUNY Buffalo)
Dept of Media & Communication, Interdisciplinary Program of Digital Convergence Business, YeungNam University, 214-1, Dae-dong, Gyeongsan-si, Gyeongsangbuk-do, South Korea, Zip Code 712-749
Editor-In-Chief of Journal of Contemporary Eastern Asia, https://jceasia.org/
Guest Editors of Social Science Computer Review, Journal of Computer-Mediated Communication, Asian Journal of Communication, Scientometrics, Quality & Quantity, Technological Forecasting & Social Change
Editorial Boards of Scientometrics, Quality & Quantity, International Journal of Internet Science, Big Data & Society, CollNet Journal of Scientometrics & Information Management, Knowledge Economy, International Journal of Technology Management & Sustainable Development, Social Media & Society, Triple Helix - A Journal of University-Industry-Government Innovation and Entrepreneurship, Technological Forecasting & Social Change, Journal of Data & Information Science, Scholarly Metrics and Analytics (a specialty of Frontiers in Library and Information Science)
Director of Cyber Emotions Research Institute
President of WATEF (World Association for Triple Helix & Future Strategy Studies,
Formerly, Asia Triple Helix Society)
Formerly, Directors of World Class University Webometrics Institute and TEDxPalgong ,Visiting Scholar of Oxford Internet Institute https://www.oii.ox.ac.uk/people/han-woo-park/ Research Associate of Royal Netherland Academy (NIWI-KNAW http://virtualknowledgestudio.nl/people/former-vks-members/
Twitter network map of #ACPC2017 1st day using NodeXLHan Woo PARK
ย
The Asian Conference for Political Communication 2017 is a biennial conference spearheaded by the Media Programme Asia of Konrad-Adenauer-Stiftung. Join fellow influencers in academia, politics and government on this special occasion. Talk about challenges on social media, discuss the odds and threats of #twiplomacy, analyze the phenomena of rising populism and watch the hottest tools for e-campaigning.
์ธ๊ณ์ฐํ๊ดํ๋ ฅ์ดํ Watef ํจ๋์ ๊ณต์งํฉ๋๋คHan Woo PARK
ย
์ธ๊ณ์ฐํ๊ดํ๋ ฅ์ดํ http://www.watef.org ํจ๋์ ๊ณต์งํฉ๋๋ค.
ํ ๋ก ์๋ก ์ฐธ์ฌ๊ฐ๋ฅํ ๋ถ๋ ๋ชจ์งํฉ๋๋ค. ๋ง์ ๊ด์ฌ์ ๋ถํ๋๋ ค์.
2017 Triple Helix Conference Special Issue Session http://www.triplehelix-korea.org/
Theme (์ฃผ์ ): Measuring Triple Helix Synergies and Innovations using Scientometric, Technometric, Informetric, Webometric, and Altmetric Data
9์ 15์ผ ๊ธ์์ผ 16:45 - 18:15 ์ฅ์: ๋๊ตฌ์์ค์ฝ
ํ์ด์ค๋ถ ๋ผ์ด๋ธ ์์ค๊ณ: ์์ธ์ (์ฆ๊ฑฐ์ด ์ฌ๋ฌผ์ธํฐ๋ท ๋ํ)
Organizers
Mi Young Chong (Univ. of North Texas, USA) miyoungchong@my.unt.edu
Han Woo Park (Yeungnam University, South Korea)
๋ฐํ์: ๊น๋ํ, Brandon Moore, ์ค์ ์
Presenter: Leo Kim
Title: From Institution to Individual: Data-Based Reflection of Triple-Helix Operation in South Korea
Biography: Leo Kim is the CEO of Ars Praxia. He graduated from LSE (Methodology, Master) and University of Sussex (Science and Technology Studies, PhD). He has been developing the methodology of semantic network analysis and has a professional background in innovation studies and unstructured data analysis.
Presenter: Brandon Moore
Title: Improving open data accessibility: Using domain driven design and microservice architecture to create user friendly open data systems
Biography: Brandon Moore is a solutions architect for Nationstar Mortgage LLC. He has 20 years experience as a software engineer and is currently the lead architect for a new reverse mortgage platform. He is also an Information Science PhD student at the University of North Texas. His research interests are text analytics and data mining.
Presenter: Jungwon Yoon
TItle: Triple Helix Dynamics of Technological Innovation Systems in South Korea: A Comparative Analysis
Biography: Jungwon Yoon is a research fellow at Soongsil University in South Korea. She received her Ph.D. in Sociology of Technology and Science from Georgia Institute of Technology, USA. Her areas of expertise are in science and technology studies (STS), including science & technology policy, sociology of sciences, innovation studies, and technology management. Her current research focuses on dynamics of innovation systems in both South and North Korea. Her latest publications include "Triple helix dynamics of South Koreaโs innovation system: a network analysis of inter-regional technological collaborations" and "Quintuple helix structure of Sino-Korean research collaboration in science."
Korean manual for nodexl fb, flickr, twitter, youtube, wiki
1. * This slide was made by Han Woo Park and his students to help
Korean users use the NodeXL
์ด ์ฌ๋ผ์ด๋๋ Marc Smith, Analyzing Social Media Networks with
NodeXL์ 3,4์ฅ์ ๊ธฐ์ด๋ก ํ๊ตญ ์ด์ฉ์๋ค์ด NodeXL์ ์ฝ๊ฒ ์ฌ์ฉํ
์ ์๋๋ก ๋ง๋ ๋งค๋ด์ผ์. NodeXL ์ต๊ทผ ๋ฒ์ ์ ์ฌ์ฉํ์ผ๋ฉฐ ์ฌ๋ก ๋
ํ ์์ ์ ์์ดํจ.
- ์์ฑ์ผ: 2011๋ 07์ 28์ผ
31. * This slide was made by Han Woo Park and his students to help Korean
users use the NodeXL
NodeXL Chapter 11: FaceBook
๋ ธ๋์์ ์ ์ด์ฉํ ํ์ด์ค๋ถ ๋คํธ์ํฌ ๋ถ์
* ์ด ์ฌ๋ผ์ด๋๋ Marc Smith, Analyzing Social Media Networks with
NodeXL์ 11์ฅ์ ๊ธฐ์ด๋ก ํ๊ตญ ์ด์ฉ์๋ค์ด ๋ ธ๋ ์์ ์ ์ฝ๊ฒ ์ฌ์ฉํ ์
์๋๋ก ๋ง๋ ๋งค๋ด์ผ์. ๋ ธ๋์์ ์ต๊ทผ ๋ฒ์ ์ ์ฌ์ฉํ์ผ๋ฉฐ ์ฌ๋ก ๋ํ ์
์ ์ ์์ดํจ.
โข์ด ๋งค๋ด์ผ์ ์ด์ฉํ ๋์๋ ๋ค์๊ณผ ๊ฐ์ด ๋ฐํ ์ฃผ๊ธฐ๋ฐ๋.
์ดํ์ง, ๊น์ง์, ๋ฐํ์ฐ (2010). ๋ ธ๋์์ ์ ์ด์ฉํ ํ์ด์ค๋ถ ๋คํธ์ํฌ ๋ถ์.
โข๊ฒฝ์ฐ: ์๋จ๋ํ๊ต.
59. Friendwheel to Pinwheel :
A Facebook Visualization the NodeXL way
โข 1๋จ๊ณ ; Reorder vertices within the clusters.
โข 2๋จ๊ณ ; convert a circle layout to a polar
layout.
โข 3๋จ๊ณ ; Turn a ring into a series of flames.
60. Facebook importer download
โข That's really surprising. For the social net importer, you should be able to
place the two files from http://socialnetimporter.codeplex.com/ in the plug-
ins directory under C:program files (x86)Social media research and then
restart nodeXL. After this is done, load the nodexl template and it should
automatically detect and present to you "Import from Facebook user's
network" under the import menu. There could be an issue with Korean
characersets, but I doubt it, since they are unicode and we have unicode all
sorted out. The latest version is on codeplex and is pretty stable. If you get it
working, you wil be impressed by the speed and accuracy.
โข
โข As for namegenweb, that has also been tested. As long as you start from
โข https://apps.facebook.com/namegenweb it should work fine.
โข
70. This slide was made by Han Woo Park and his students to help Koreans to use the NodeXL
๋ ธ๋์์ ์ ์ด์ฉํ ํ๋ฆฌ์ปค ๋คํธ์ํฌ ๋ถ์
์ด ์ฌ๋ผ์ด๋๋ Marc Smith, Analyzing Social Media Networks with NodeXL์
13์ฅ์ ๊ธฐ์ด๋ก ํ๊ตญ ์ด์ฉ์๋ค์ด ๋ ธ๋์์ ์ ์ฝ๊ฒ ์ฌ์ฉํ ์ ์๋๋ก ๋ง๋ ๋งค๋ด
์ผ์. ๋ ธ๋์์ ์ต๊ทผ ๋ฒ์ ์ ์ฌ์ฉํ์ผ๋ฉฐ ์ฌ๋ก ๋ํ ์์ ์ ์์ดํจ.
โข์ด ๋งค๋ด์ผ์ ์ด์ฉํ ๋์๋ ๋ค์๊ณผ ๊ฐ์ด ๋ฐํ ์ฃผ๊ธฐ๋ฐ๋.
์์ , ๋ฐํ์ฐ(2010). ๋ ธ๋์์ ์ ์ด์ฉํ ํ๋ฆฌ์ปค ๋คํธ์ํฌ ๋ถ์
โข๊ฒฝ์ฐ: ์๋จ๋ํ๊ต
101. * This slide was made by Han Woo Park and his students to help Koreans
to use the NodeXL
NodeXL Chapter 10: Twitter
๋ ธ๋์์ ์ ์ด์ฉํ ํธ์ํฐ ๋คํธ์ํฌ ๋ถ์
* ์ด ์ฌ๋ผ์ด๋๋ Marc Smith, Analyzing Social Media Networks
with NodeXL์ 10์ฅ์ ๊ธฐ์ด๋ก ํ๊ตญ ์ด์ฉ์๋ค์ด ๋ ธ๋ ์์ ์ ์ฝ๊ฒ
์ฌ์ฉํ ์ ์๋๋ก ๋ง๋ ๋งค๋ด์ผ์. ๋ ธ๋์์ ์ต๊ทผ ๋ฒ์ ์ ์ฌ์ฉํ์ผ
๋ฉฐ ์ฌ๋ก ๋ํ ์์ ์ ์์ดํจ.
105. *Twitter
@replies and@mentions Retweet
ํธ์ํฐ์์ ์๋ก๊ฐ์ ๋๋๋ ๋ํ์ ๋ค๋ฅธ ์ฌ๋์ ํธ์์ ๋์ํ๊ฑฐ๋ ๋
๋ฐฉ์. ํธ์์ ์์์ @user`s name ํ ๋ค๋ฅธ ์ฌ๋(๋์ ํ๋ก์)์๊ฒ ์๋ ค
๋ฉด reply๋ก ์ธ์. ํธ์ ์ฌ์ด์ @user`s ์ฃผ๊ณ ์ถ์ ํธ์์ ์ ํ ๋ ์ฌ์ฉ.
name์ด ๋ค์ด๊ฐ๋ฉด mention์ผ๋ก ์ธ์ํจ.
tweet starts off with โRT @ASAnews.โ RT
- @ebertchicago: I was just reading in John stands for โretweet,โ and is followed by an
Waters' new book "Role Modelsโ @mention of the ASAnews account
- I was just reading in John Waters' new book
"Role Modelsโ @ebertchicago how about it? *๋ชจ๋ RT๋ ๋ชจ๋ @reply ๋ฅผ ํฌํจํ์ง๋ง, ๋ชจ
๋ @reply๊ฐ ๋ชจ๋ RT๋ฅผ ํฌํจํ์ง๋ ์์.
*๋ชจ๋ @replies๋ ๋ชจ๋ @mentions, ๊ทธ๋ฌ๋ ๋ชจ
๋ @mentions์ ๋ชจ๋ @replies๊ฐ ์๋.
#robotpickuplines โIf I could rearrange the
#Hashtag qwerty keyboard, I'd put u and i ..
ํ ๊ฐ์ง ์ฃผ์ ๋ก ์ด์ผ๊ธฐํ ๋ ๊ฒ์ํ๊ธฐ ์ฝ๊ฒ ํด์ฃผ๋ oh, wait, nevermindโ
ํธ์ํฐ ๊ณ ์ ์ ํ๊ทธ. ์ฌ๋๋ค์ ๊ณตํต์ ๊ด์ฌ์ฌ๋ฅผ
ํํํ๋ค.
126. *Twitter
REST API and Whitelisting an account
โข Representational State Transfer (REST) Application
Programming Interface (API) are used by Twitter to
provide data in XML or JSON to third party clients like
TweetDeck, Twhirl, and also NodeXL
โข Regular account is limited to 150 queries per hour.
โข For data intensive tasks, one might need to whitelisting
his/her account.
127. *Twitter
Whitelisting an account
โข To do this visit:
โ http://twitter.com/help/request_whitelisting
โ Fill in the form and once whitelisted use the ID into NodeXL
Twitter import interface.
128. This slide was made by Han Woo Park and his students to help Korean users use the NodeXL
๋ ธ๋์์ ์ ์ด์ฉํ ์ ํ๋ธ ๋คํธ์ํฌ ๋ถ์
์ด ์ฌ๋ผ์ด๋๋ Marc Smith, Analyzing Social Media Networks with NodeXL์
14์ฅ์ ๊ธฐ์ด๋ก ํ๊ตญ ์ด์ฉ์๋ค์ด ๋ ธ๋์์ ์ ์ฝ๊ฒ ์ฌ์ฉํ ์ ์๋๋ก ๋ง๋ ๋งค๋ด
์ผ์. ๋ ธ๋์์ ์ต๊ทผ ๋ฒ์ ์ ์ฌ์ฉํ์ผ๋ฉฐ ์ฌ๋ก ๋ํ ์์ ์ ์์ดํจ.
โข์ด ๋งค๋ด์ผ์ ์ด์ฉํ ๋์๋ ๋ค์๊ณผ ๊ฐ์ด ๋ฐํ ์ฃผ๊ธฐ๋ฐ๋.
์์ , ๋ฐํ์ฐ(2010). ๋ ธ๋์์ ์ ์ด์ฉํ ์ ํ๋ธ ๋คํธ์ํฌ ๋ถ์
โข๊ฒฝ์ฐ: ์๋จ๋ํ๊ต
159. Contents
1.Key Features of Wiki Systems
2.Wiki Networks from Edit Activity
3.Identifying Different Types of Editors within a Wiki
Project
4.NodeXL Visualization Strategies for Revealing
Distinct User Types
5.Identifying High-Quality Contributors in Article Talk
Pages
6.Navigating Lostpedia: Using NodeXL to Reveal the
Large-Scale Collaborative Structure of Wiki
systems
160. โwikiโ means โQuickโ in Hawaiian
Ward Cunningham invented WikiWikiWeb
1995 to allow a group to easily and quickly
edit a set of web pages without having to
know HTML or deal with moving files back
and forth to a web server.
--๏ knowledge repositories
161. Tree different types of questions
from NodeXL
1.Study a set of wiki pages at the Empire Wiki that
are related by the Castle Project, and it seeks to
identify different types of contributors to that
project based on both their network attributes and
key variables related to the types of pages they do,
and do not, edit. ?
2.The quality of online discussion on the โtalkโ
3. Revealing Large-scale structure of editing patterns
in wikis, drawing on data from Lostpedia(http://en.
wikipedia.org/wiki/Lostpedia).
Lostpedia
162. KEY FEATURES OF WIKI
Chapter 15
SYSTEMS
This article page from the English-language Wikipedia displays content
and illustrates discussion, edit, and history tabs. These tabs are standard
to most wiki systems and they provide access to edit records from which
edge relationships and attributes can be measured.
Copyright ยฉ 2011, Elsevier Inc. All rights Reserved
163. Chapter 15
Wiki pages have a related history page that depicts the timing of every
edit, indicates the editor or IP address responsible for the edit, provides
space for a brief description of the edit, and displays links to the state of
the page before and after the edit. History pages are important sources of
network and attribute data in wiki systems.
Copyright ยฉ 2011, Elsevier Inc. All rights Reserved
164. KEY FEATURES OF WIKI
Chapter 15
SYSTEMS
This article talk page is used to coordinate decisions about the best
contents for the article page. The edits to this page are made by people
who have an interest in the content page and are often made by people
who actively edit the article page. This page shows evidence both of
content-based discussion and the implementation of templates to
encourage compliance with community editing norms.
Copyright ยฉ 2011, Elsevier Inc. All rights Reserved
165. KEY FEATURES OF WIKI
Chapter 15
SYSTEMS
This page reports a partial history of edits made by a wiki user. These
contribution pages are an important source of information about editors.
This image also shows a drop-down menu with a range of page types or
โnamespacesโ in Wikipedia and typical to many wikis. The tendency of
editors to edit pages in certain namespaces and not others provides
important clues about the roles they play in the wiki community.
Copyright ยฉ 2011, Elsevier Inc. All rights Reserved
167. Chapter 15
History of the Project Castle page
This study of wiki social networks used the full revision history of the
Project Castle page in the Empire Wiki as both a definition of the
community of interest and as a source of user IDs. We were interested in
the roles played within the community of contributors to these pages.
Therefore, when we scraped all of these history pages, we were sure to
get all active contributors to this project. Starting from a list of URLs for
Project history pages, the web scraping software returns an Excel sheet
populated with all text that occurs after the edit date and prior to the (talk &
Contribs) link.
Copyright ยฉ 2011, Elsevier Inc. All rights Reserved
168. WIKI NETWORKS FROM EDIT
ACTIVITY
โข Many interesting ways to analyze Wikipedia based on
the history of activity and interaction of its users
โข
Carter Butts raised several foundational issues related
to the challenge of interpreting activity data into a
network representation
Networks are composed of vertices or entities that are
connected through edges that represent the
relationships between them. Both vertices and
relationships can have attributes, such as the strength
of a tie between vertices or the length of time a vertex
has been part of the network.
169. WIKI NETWORKS FROM EDIT
ACTIVITY
โข vertex =Each distinct user account
โข An edge= one of many activities that
display some type of interaction between
two users
170. Identifying different types of editors within a wiki project
Network Vertices Edges Weighted Directed
Page Link Network Pages Hyperlinks Yes or No Yes
User Talk Page(ig, Users Comments on another userโs Yes Yes
profile)Network profile page(eg,user talk
page)
User Discussion Users Comments posted in reply to Yes Yes
Network each other on an Article
Discussion page
User to Page Pages and User edits per page Yes No
Affiliation Network users
Page Co-editor Pages Co-editors Yes No
Network
User Co-edit Users Co-edited pages Yes No
Network
Category network Categories Shared pages Yes No
Project Network Projects Shared pages or shared Yes No
members
Several Primary Types of Wiki Networks That Can be Derived from Edit Records
171. Wiki Social Network Sampling
Frame and Data Collection
1. Constructed a list of URLs of history
pages for every article related to โproject
Castel,โ as tagged by users.
2. A commercial web scraping program was
used to generate an Excel spreadsheet
containing a list of each user making an
edit to each respective article during the
sample period(about 7months)
172. Defining Edges and Attributes in
Wiki Social networks
โข One editor wanted to contact another editor
outside the context of the specific project pages
ex)a directed edge form vertex A to vertex B
represents user A making an edit on the talk page
of user B
Two types of vertex attributes
1) A set of attributes describing the structural
position of each vertex
2) A set of attributes generated from measures of
participation in the Empire Wiki community and
participation in
Wiki Network Data Collection
173. Wiki Network Data Collection
โข To obtain these data, we started with the
list of sampled users in Excel.We then
used the Web scraper to go through the
history page of each sampled user and
build an Excel spreadsheet with the name
of the user whose page was being
scraped. The name of each user making
an edit and the time stamp for each edit.
174. NODEXL VISUALIZTION STRATEGIES
FOR REVEALING DISTINCT USER TYPES
illustrate how NodeXL can be used to analyze
larger chunks of network data from wiki sites
1) Construct a graph of the overall network
2) Visually represent different vertex attributes
3) Search for structural similarities among
individuals exhibiting similar behaviors or
occupying similar roles.
175. Chapter 15
NodeXL uses spreadsheet columns to store attributes of each vertex and
can be transformed using standard Excel formulas. In this case, we see a
sample of some Empire Wiki editorsโ overall activity and the proportion of
pages that they edited that were related to Project Castle.
Copyright ยฉ 2011, Elsevier Inc. All rights Reserved
176. Chapter 15
NodeXL allows you to assign gradients of vertex colors that correspond
with data attributes in the spreadsheet. This helps make the resulting
graph easier to read and analyze and highlights key features of interest.
Copyright ยฉ 2011, Elsevier Inc. All rights Reserved
177. Chapter 15
This NodeXL wiki network graph shows a well defined outer ring of users
and a strong inner core. Only a handful of vertices connect the outer ring
to the inner core. Without these nodes, the population would be highly
fragmented.
Copyright ยฉ 2011, Elsevier Inc. All rights Reserved
178. Chapter 15
The NodeXL wiki network on the left displays the relative proportions of Project Castle
edits among users sampled. Dark green indicates the lowest proportion of edits, and light
green is the highest. The figure on the right displays the volume of edits to the usersโ
respective user pages. Dark blue indicates the lowest edit volume, and light blue
represents the highest edit volume. Users who connect the outer ring to the inner core in
the previous visualization have few Project Castle edits, and those users who display a
high volume of edits are relatively isolated in the previous visualization. This indicates that
Project Castle is not strongly connected to the larger Empire Wiki community.
Copyright ยฉ 2011, Elsevier Inc. All rights Reserved
179. Chapter 15
This figure compares the degree 1.5 ego network graphs of four different exemplary
types of Project Castle contributors. Ego network graphs with automated layouts are
good ways to identify potential structural signatures of online roles. In this instance, we
see evidence that system administrators tend to have more connection to others
involved in the project than do the actual substantive experts. Interestingly, for both
sysops and substantive contributors, the higher-level contributors tend to have fewer
connections.
Copyright ยฉ 2011, Elsevier Inc. All rights Reserved
180. 1)Making Top Wiki editors Stand Out by
Visually Formatting the Network Graph
2)Interpreting Wiki Network Graphs for
Evidence of Distinctive Social Roles
3) Using Subgraph Images to Distinguish
between User types
4) Seeing the trees and Forest with Wiki
Network Analysis
181. IDENTIFYING High-quality
contributors in article talk pages
1) Tasks and Strategies for Identifying Types
of Contributors by Visualizing Article
Discussion Page Networks
2) Searching for Structural Signatures of
Confrontation and Deliberation in Wiki
Article Talk Page Networks
182. Chapter 15
NodeXL can make use of the full range of Excel 2007 features, for
example, using an โif-statementโ to assign vertex color according to a
categorical defi nition of low, medium, and high. A categorical assignment
like this one is used to highlight large differences in the measured
attribute. In this case, we can concentrate on the difference between
contributors who are actively improving the quality of the discussion
(green) from those who are actively undermining it (red).
Copyright ยฉ 2011, Elsevier Inc. All rights Reserved
183. Chapter 15
This NodeXL network graph depicts user-to-user talk page connections from a
Wikipedia policy article. The graph illustrates one way that styles of contribution are
tied to structural attributes. Note that the red nodes (most confrontational) are
involved in the strongest dyadic ties, and they tend to have the highest outdegree. In
contrast, the most deliberative contributors tend to have fewer partners and do not
necessarily involve themselves in intense dyadic interactions. Observations like
these can provide direction for further research that statistically tests the strength of
these observer relations. Ultimately, if those measures are robust predictors, they
could be used in automated systems for identifying more or less collaborative
contributors, assessing community health, and deciding where interventions or
support might be most helpful.
Copyright ยฉ 2011, Elsevier Inc. All rights Reserved
184. NAVIGATING LOSTPEDIA: USING NODEXL TO
REVEAL THE LARGE-SCALE COLLABORATIVE
STRUCTURE OF WIKI SYSTEMS
1)Creating an Overview Network Map of
Lostpedia Content in Node XL
2)Creating an Overview Map of Lostpedia
Users
3)Moralizing Data to Infer Stronger
Connections
185. Chapter 15
Lostpediaโs article about the Statue of Taweret with links to its associated
Discussion and Theory pages. Similar to other wiki systems, Lostpedia include
links to History pages and an Edit page. The Theory page is an additional type
of page for contributor interpretations of what is happening and why, whereas
the articles are more descriptive of what occurred in the show.
Copyright ยฉ 2011, Elsevier Inc. All rights Reserved
186. Chapter 15
NodeXL Lostpedia wiki page-to-page co-edit network visualization and
Vertex worksheet showing only those pages with more than 50 co-editors.
All types of pages were considered, but only Article pages
(maroon), Discussion pages (orange), Theory pages (green), and User
Talk pages (deep pink) were co-edited enough to show up. The Harel-
Koren Fast Multiscale Layout identifies natural groupings such as the main
cluster of articles and the cluster of interrelated Theory pages. Size is
based on total user edits of a page, and opacity is based on degree.
Subgraph images show small dense clusters for the displayed vertices.
Copyright ยฉ 2011, Elsevier Inc. All rights Reserved
187. Chapter 15
NodeXL visualization of Lostpedia wiki user-to-user affiliation network
connecting users (vertices) based on the number of unique pages they
have both edited (weighted edges). Two types of edges are included:
those connecting users based on co-edits of 20 or more Theory pages
(green) and those connecting users based on co-edits of 150 or more
articles (maroon). Vertex size is based on total wiki edits, and color is
based on the percentage of pages that are Theory pages (green vertices
edit mostly Theory pages and maroon vertices edit mostly Article pages).
Boundary spanners and important individuals are easily identified.
Copyright ยฉ 2011, Elsevier Inc. All rights Reserved
188. Chapter 15
NodeXL Edges worksheet and visualization of a Lostpedia wiki user-to-
user affiliation network graph with edges filtered based on the number of
pages that users share as a percentage of the total number of edited
pages. The number of edges for frequent editors like Santa (highlighted in
red) are significantly reduced in the graph, but size indicates that they exist
with those filtered out of the graph.
Copyright ยฉ 2011, Elsevier Inc. All rights Reserved
189. DATA COLLECTION FROM WIKI
SYSTEMS
โข Data collection from wikis is not automatic.
โข Data collection from wikis requires a
combination of technical skill and effort
from the Empire wiki
โข Second example extracted data directly
from Wikipedia and required no special
tools
190. PRACTITIONERโS SUMMARY
โข Wikis are complex social media systems that
give rise to many types of relationships
โข The complexity inherent in wiki systems is the
source of both challenge and opportunity for
practitioners.
โข Wikis can provide valuable insights because
they are places where collaboration happens
and value is created through informal
organization
191. RESEARCHERโS AGENDA
โข Node XL as well as browser-based
network visualization tools like Touch
Graph are helping expand participation in
social network analysis .
โข Wikis are rich settings in which to study
the dynamics of diffusion
192. Analyzing Social Media Networks with NodeXL
Insights from a Connected World
Chapter 15
Wiki Networks
Connections of Creativity and Colla
boration
Thank you
Copyright ยฉ 2011, Elsevier Inc. All rights
Reserved 192