Bruno Goncalves1;2, Nicola Perra1;3, and Alessandro Vespignani1;2;4
1Center for Complex Networks and Systems Research,
School of Informatics and Computing, Indiana University, IN 47408, USA
2Pervasive Technology Institute, Indiana University, IN 47404, USA
3Linkalab, Complex Systems Computational Lab. - 09100 Cagliari Italy and
4Institute for Scientic Interchange, Turin 10133, Italy
Maio 2011
This document discusses the possibilities and pitfalls of using social networking tools in educational settings. It begins by providing background on the prevalence of social networking use among teens and adults. It then explores the capabilities of social networks, including uses for marketing, knowledge sharing, and connecting with mentors. The document outlines how social network algorithms work to suggest connections. It acknowledges both the potential benefits of using networks like Edmodo in classrooms, while also warning of dangers like cyberbullying. It concludes by advocating for setting clear policies on appropriate social network use in schools.
An overview of Web research areas of interest to social scientists presented at Brunel University 3 March 2010, including an overview of my attempts to understand social influence online for my PhD thesis (http://alekskrotoski.com/tags/phd). includes general findings and an overview of the themes discussed in BBC2's Virtual Revolution series.
Pratt SILS Knowledge Organization Spring 2010PrattSILS
This document discusses subject access for visual materials such as artwork, maps, and photographs. It notes that subject headings provide an important third discovery point for library users beyond keyword and field searching. However, applying subject headings to non-textual materials poses unique challenges. For artwork, interpretations can vary, and metadata standards like ARTstor allow for inconsistent subject descriptions. Maps also require careful assessment to determine if geographic labels constitute subject headings. The document examines these issues and debates around subject access for non-textual library materials.
Authors: Philip E. Banyard, Jean Underwood
There are concerns that the Internet has created new risks for our society and in particular for young people. We argue that the way we frame these risks is affected by the way we view young people, in particular their maturity and ability to make choices for themselves.
Jana Diesner, "Words and Networks: Considering the Content of Text Data for N...summersocialwebshop
The document discusses extracting network data from text documents. It outlines extracting named entities like people, places, events as nodes and linking them based on proximity, syntax, and statistics to build networks. Examples are provided of networks constructed from news articles about Sudan showing prominent individuals and organizations and how their centrality changes over time. The goal is to understand socio-technical networks and their co-evolution with knowledge and structure from large-scale text data.
This literature survey discusses papers on the topological structure of social networks and information propagation. Regarding network structure, papers found that social networks like Facebook exhibit scale-free and small-world properties with high clustering. Different networks may have different structures depending on factors like symmetry. Regarding information spread, factors like sender involvement, tie strength, and communication influence forwarding. Messages tend to spread through closely connected friendship networks rather than broadly. Key influencers and the structure of interaction graphs also impact propagation patterns.
Jessica Vitak, "When Contexts Collapse: Managing Self-Presentation Across Soc...summersocialwebshop
This document discusses challenges and strategies for researching the complex relationships between technology, privacy, and interpersonal connections. It summarizes key points in 3 sentences:
The document outlines challenges in defining research questions, collecting representative data, and analyzing relationships when studying how people navigate privacy and disclosure across overlapping social contexts online. It explores how the merging of distinct audiences onto single social media platforms can influence users' sharing behavior and perceived social capital. Theoretical frameworks are discussed for understanding how privacy controls, audience characteristics, and self-disclosure impact users' social capital within their networks.
This document discusses the possibilities and pitfalls of using social networking tools in educational settings. It begins by providing background on the prevalence of social networking use among teens and adults. It then explores the capabilities of social networks, including uses for marketing, knowledge sharing, and connecting with mentors. The document outlines how social network algorithms work to suggest connections. It acknowledges both the potential benefits of using networks like Edmodo in classrooms, while also warning of dangers like cyberbullying. It concludes by advocating for setting clear policies on appropriate social network use in schools.
An overview of Web research areas of interest to social scientists presented at Brunel University 3 March 2010, including an overview of my attempts to understand social influence online for my PhD thesis (http://alekskrotoski.com/tags/phd). includes general findings and an overview of the themes discussed in BBC2's Virtual Revolution series.
Pratt SILS Knowledge Organization Spring 2010PrattSILS
This document discusses subject access for visual materials such as artwork, maps, and photographs. It notes that subject headings provide an important third discovery point for library users beyond keyword and field searching. However, applying subject headings to non-textual materials poses unique challenges. For artwork, interpretations can vary, and metadata standards like ARTstor allow for inconsistent subject descriptions. Maps also require careful assessment to determine if geographic labels constitute subject headings. The document examines these issues and debates around subject access for non-textual library materials.
Authors: Philip E. Banyard, Jean Underwood
There are concerns that the Internet has created new risks for our society and in particular for young people. We argue that the way we frame these risks is affected by the way we view young people, in particular their maturity and ability to make choices for themselves.
Jana Diesner, "Words and Networks: Considering the Content of Text Data for N...summersocialwebshop
The document discusses extracting network data from text documents. It outlines extracting named entities like people, places, events as nodes and linking them based on proximity, syntax, and statistics to build networks. Examples are provided of networks constructed from news articles about Sudan showing prominent individuals and organizations and how their centrality changes over time. The goal is to understand socio-technical networks and their co-evolution with knowledge and structure from large-scale text data.
This literature survey discusses papers on the topological structure of social networks and information propagation. Regarding network structure, papers found that social networks like Facebook exhibit scale-free and small-world properties with high clustering. Different networks may have different structures depending on factors like symmetry. Regarding information spread, factors like sender involvement, tie strength, and communication influence forwarding. Messages tend to spread through closely connected friendship networks rather than broadly. Key influencers and the structure of interaction graphs also impact propagation patterns.
Jessica Vitak, "When Contexts Collapse: Managing Self-Presentation Across Soc...summersocialwebshop
This document discusses challenges and strategies for researching the complex relationships between technology, privacy, and interpersonal connections. It summarizes key points in 3 sentences:
The document outlines challenges in defining research questions, collecting representative data, and analyzing relationships when studying how people navigate privacy and disclosure across overlapping social contexts online. It explores how the merging of distinct audiences onto single social media platforms can influence users' sharing behavior and perceived social capital. Theoretical frameworks are discussed for understanding how privacy controls, audience characteristics, and self-disclosure impact users' social capital within their networks.
2009 - Connected Action - Marc Smith - Social Media Network AnalysisMarc Smith
Review of social media network analysis of Internet social spaces like twitter, flickr, email, message boards, etc. Network analysis and visualization of social media collections of connections.
This document discusses the design of data for science. It addresses two topics: (1) science through platforms for mediated communication like social media platforms, and (2) science on social science contributions about fundamentals of collaboration. For topic 1, it notes issues with awareness, content sharing, coordination, and longevity on current platforms. For topic 2, it emphasizes the importance of data quality, stability, and social responsibility for science. The document calls for addressing the design of data for science directly through interdisciplinary collaboration between designers, statisticians, computer scientists, data scientists, and social scientists.
First presented at Future of Digital Identity, British Library, 7 Jan 2010. Updated for presentation at Privacy and the Law Conference, London, 1 Dec 2010, and for an OxPILS seminar at Balliol College, 23 May 2012.
This document promotes a social intranet solution from Jive Software. It summarizes that traditional intranets focused only on content without enabling collaboration, but social intranets add social features like profiles, groups, activity streams. This turns employees into information curators and addresses challenges like inefficient searching and information silos. Social intranets drive cross-department collaboration and have proven successful for customers like Alcatel-Lucent, Bupa, and CSC in boosting productivity, engagement, and information sharing. Jive offers an all-in-one social intranet solution with features for mobile access, Outlook integration, and prescriptive services.
The document discusses Members of Congress' use of Twitter to communicate with constituents. It notes that while Members have traditionally used conservative communication methods, Twitter allows for more interactive exchanges. The document analyzes the first year of tweets from Members and finds they mainly use Twitter to distribute vetted content and promote appearances, but not for interactive conversations. It also discusses the potential for Twitter to democratize interactions if used innovatively. Restrictions on Members' communication and the adoption of new technologies are also examined.
This document summarizes a study that compares the use of Facebook Groups and Causes as tools for cyberactivism in Chile and the Concepción area. The study aimed to determine if these applications are used for cyberactivism related to real-world issues and places in Chile. It also sought to identify trends in topics of discussion and whether there are correlations between topics discussed nationally and locally. The main findings were that Facebook is mainly used to strengthen existing social ties rather than create new connections, and that it allows for distributed organization without centralized control of information.
2009-HICSS-42-Best paper-A conceptual and operational definition of ‘social r...Marc Smith
1) The document defines social roles as cultural objects that organize behavior and structure social positions through shared understandings of acceptable actions.
2) It provides a conceptual definition of social roles as patterns of social, structural, and behavioral attributes that can be identified through content analysis and statistical analysis of social networks and behaviors.
3) The paper demonstrates applying this combined conceptual and operational definition to identify key social roles in Usenet and Wikipedia through qualitative and quantitative analysis.
This document discusses the emergence of tools and practices to help people manage the growing amount of information and data. It describes how data visualization tools will play an important role, allowing people to interact with and find patterns in large datasets. These tools will include network diagrams, interactive visualizations that allow user comments and sharing, and visualizations created by foundations to communicate data to broad audiences. The document also notes that social filtering, ambient displays, agents and interfaces will be other important tools to help people cope with information overload in the coming decade.
This presentation was developed to help a client address best practices for building an online community within the workplace. It was based upon a great deal of research and study of the topic and should help those who are seeking information or wish to start an online community, as it pulls together a great amount of data and resources on the topic.
This document outlines a thematic unit on technology and communication that uses the short story "From Email to Bill Gates" by John Seabrook. Over the course of 12 days, students will explore the story, learn about key figures like Bill Gates and Steve Jobs, discuss different forms of social media, and compare/contrast the work and impact of Gates and Jobs. Assessment includes sending an email to Gates, answering text questions, debating topics, and a final exam on the story content and themes. The goal is for students to understand how technology facilitates communication while developing skills like critical thinking.
Crisis Information Management in the Web 3.0 AgeAxel101
The document discusses crisis information management in the Web 3.0 age. It proposes combining social and semantic web data to provide structured and relevant information to enhance situational awareness for emergency responders. A process is described that involves collecting information from citizens and social media, classifying it, enriching it with linked open data, and providing it to decision makers to improve response. Future work could involve user studies and applying machine learning to the classification process.
2010-November-8-NIA - Smart Society and Civic Culture - Marc SmithMarc Smith
This document discusses how social media and social networks are enabling new forms of civic participation and collective action. It notes that citizens are increasingly using social media to find government services, engage in discussions, and measure public opinion. The document also discusses how social network analysis can be used to analyze patterns in social media networks and identify influential users. It provides an overview of various social media platforms and the types of social networks and connections that exist within them.
Social computation of emergent networks on user generated contentMarkus Strohmaier
This document summarizes a presentation on social computational systems given by Markus Strohmaier at the GI Workshop on Web Science in 2010. It discusses the rise of user-generated content and social networks. It also explains that social computational systems emerge from the complex interactions between people and computers on online platforms. Properties of these systems like findability, utility, navigability and relevance can be influenced by social computation processes. The presentation agenda discusses social computational systems, navigability, semantics and the future of these systems. An example is given of how the connectivity of the web is shaped by social mechanisms like preferential attachment.
Digital Trails Dave King 1 5 10 Part 1 D3Dave King
This document provides an overview and agenda for a tutorial on extracting intelligence from digital traces and trails left by web and mobile users. It discusses the proliferation of digital devices that create extensive data about people's online and mobile activities. Examples are given of different types of digital traces, including cookies, web bugs, location data, and social media interactions. Concerns about privacy are also mentioned as vast amounts of personal data are now collected and analyzed.
This document discusses how Twitter can be used to develop a learning community. It provides background on Twitter, noting that it allows rapid dissemination of information to a network of followers. The document then discusses how Twitter can foster informal learning beyond the classroom and help create a community of learners. Some specific ways Twitter has been used for teaching include student notice boards, one-on-one support, summarizing articles, and sharing research links. The document also notes some ethical and legal issues to consider when using Twitter for teaching, such as intellectual property and maintaining student confidentiality. Overall, it argues that when guided properly, Twitter shows potential for enhancing blended learning and developing collaborative learning communities.
This document provides an agenda and background information for a Meet 'N' Greet social event on social computing. The agenda includes a welcome, getting to know each other through brief presentations, and discussing next steps and ongoing activities. The background explains the goals of building an interdisciplinary community and collaboration around social computing research across various University of Minnesota departments and disciplines. Brief biographies of committee members are also included, describing their relevant research interests.
IJERA (International journal of Engineering Research and Applications) is International online, ... peer reviewed journal. For more detail or submit your article, please visit www.ijera.com
Future of the Internet Predictions March 2014 PIP ReportVasily Ryzhonkov
This report is the latest research report in a sustained effort throughout 2014 by the Pew Research Center to mark the 25th anniversary of the creation of the World Wide Web by Sir Tim Berners-Lee. He wrote a paper on March 12, 1989 proposing an “information management” system that became the conceptual and architectural structure for the Web. He eventually released the code for his system — for free — to the world on Christmas Day in 1990. It became a milestone in easing the way for ordinary people to access documents and interact over the Internet — a system that linked computers and that had been around for years.
The Web became a major layer of the Internet. Indeed, for many, it became synonymous with the Internet, even though that is not technically the case. Its birthday offers an occasion to revisit the ways it has made the Internet a part of Americans’ social lives.
Our first report tied to the anniversary looked at the present and the past of the Internet, marking its strikingly fast adoption and assessing its impact on American users’ lives. This report is part of an effort by the Pew Research Center’s Internet Project in association with Elon University’s Imagining the Internet Center to look at the future of the Internet, the Web, and other digital activities. This is the first of eight reports based on a canvassing of hundreds of experts about the future of such things as privacy, cybersecurity, the “Internet of things,” and net neutrality. In this case we asked experts to make their own predictions about the state of digital life by the year 2025. We will also explore some of the economic change driven by the spectacular progress that made digital tools faster and cheaper. And we will report on whether Americans feel the explosion of digital information coursing through their lives has helped them be better informed and make better decisions.
This report is a collaborative effort based on the input and analysis of the following individuals.
THE RELATIONSHIP BETWEEN THE USE OF BLACKBERRY WITH THE STUDENTS’ DEMAND FULF...cscpconf
The communication technology mainly Blackberry enables a medium to facilitate mediated interpersonal communication because of its interactive ability. This aspect creates some
easiness. In the interpersonal communication keeping apart with a distance, the interactivity of the convergent media has been over the potential ability of a feedback since a person accessing
a convergent medium directly gives a feedback of the message conveyed. Blackberry as a result of an advanced technology development has been growing so fast in this life.
Based on the background and phenomenon mentioned, this research studied about “The Relationship between The Use of Blackberry with The Demand Fulfillment and Personality of
The Junior High Students in Bandung.” The research was aimed at finding out: (1) the correlation between ten intensity of blackberry use with the cognitive and affective fulfillment of
The Junior High School Students, (2) the correlation between the intensity of blackberry use with the Tense Release of The Junior High School Students, (3) the correlation between the
intensity of blackberry use with the Personal Integrative of The Junior High School Students; (4) the correlation between the intensity of blackberry use with the Socially Integrative of The
Junior High School Students; (5) the correlation between the intensity of blackberry use with the Confidence of The Junior High School Students; (6) the correlation between the intensity of blackberry use with the Tolerance of The Junior High School Students; (7) the correlation between the intensity of blackberry use with the Whole Fulfillment of The Junior High School
Students; (8) the correlation between the intensity of blackberry use with the Personality as a whole of The Junior High School Students The research used a quantitative approach with the explanatory survey method. The Theories used were: Cognitive Psychology, Technology Determinism, and Uses and Gratification . The population of the research was The Junior High School students. By using random sampling
technique, it was taken 5 schools and 200 students as the sample. The data were taken through questionnaires. The data obtained were analyzes by using the statistical test of correlation. The results of the research were shown in the forms of Bar Chart.
THE RELATIONSHIP BETWEEN THE USE OF BLACKBERRY WITH THE STUDENTS’ DEMAND FULF...csandit
The communication technology mainly Blackberry enables a medium to facilitate mediated
interpersonal communication because of its interactive ability. This aspect creates some
easiness. In the interpersonal communication keeping apart with a distance, the interactivity of
the convergent media has been over the potential ability of a feedback since a person accessing
a convergent medium directly gives a feedback of the message conveyed. Blackberry as a result
of an advanced technology development has been growing so fast in this life.
Based on the background and phenomenon mentioned, this research studied about “The
Relationship between The Use of Blackberry with The Demand Fulfillment and Personality of
The Junior High Students in Bandung.” The research was aimed at finding out: (1) the
correlation between ten intensity of blackberry use with the cognitive and affective fulfillment of
The Junior High School Students, (2) the correlation between the intensity of blackberry use
with the Tense Release of The Junior High School Students, (3) the correlation between the
intensity of blackberry use with the Personal Integrative of The Junior High School Students;
(4) the correlation between the intensity of blackberry use with the Socially Integrative of The
Junior High School Students; (5) the correlation between the intensity of blackberry use with
the Confidence of The Junior High School Students; (6) the correlation between the intensity of
blackberry use with the Tolerance of The Junior High School Students; (7) the correlation
between the intensity of blackberry use with the Whole Fulfillment of The Junior High School
Students; (8) the correlation between the intensity of blackberry use with the Personality as a
whole of The Junior High School Students
The research used a quantitative approach with the explanatory survey method. The Theories
used were: Cognitive Psychology, Technology Determinism, and Uses and Gratification . The
population of the research was The Junior High School students. By using random sampling
technique, it was taken 5 schools and 200 students as the sample. The data were taken through
questionnaires. The data obtained were analyzes by using the statistical test of correlation. The
results of the research were shown in the forms of Bar Chart.
Towards Decision Support and Goal AchievementIdentifying Ac.docxturveycharlyn
Towards Decision Support and Goal Achievement:
Identifying Action-Outcome Relationships From Social
Media
Emre Kıcıman
Microsoft Research
[email protected]
Matthew Richardson
Microsoft Research
[email protected]
ABSTRACT
Every day, people take actions, trying to achieve their per-
sonal, high-order goals. People decide what actions to take
based on their personal experience, knowledge and gut in-
stinct. While this leads to positive outcomes for some peo-
ple, many others do not have the necessary experience, knowl-
edge and instinct to make good decisions. What if, rather
than making decisions based solely on their own personal
experience, people could take advantage of the reported ex-
periences of hundreds of millions of other people?
In this paper, we investigate the feasibility of mining the
relationship between actions and their outcomes from the
aggregated timelines of individuals posting experiential mi-
croblog reports. Our contributions include an architecture
for extracting action-outcome relationships from social me-
dia data, techniques for identifying experiential social media
messages and converting them to event timelines, and an
analysis and evaluation of action-outcome extraction in case
studies.
1. INTRODUCTION
While current structured knowledge bases (e.g., Freebase)
contain a sizeable collection of information about entities,
from celebrities and locations to concepts and common ob-
jects, there is a class of knowledge that has minimal cov-
erage: actions. Simple information about common actions,
such as the effect of eating pasta before running a marathon,
or the consequences of adopting a puppy, are missing. While
some of this information may be found within the free text of
Wikipedia articles, the lack of a structured or semi-structured
representation make it largely unavailable for computational
usage. With computing devices continuing to become more
embedded in our everyday lives, and mediating an increasing
degree of our interactions with both the digital and physical
world, knowledge bases that can enable our computing de-
vices to represent and evaluate actions and their likely out-
comes can help individuals reason about actions and their
Permission to make digital or hard copies of all or part of this work for personal or
classroom use is granted without fee provided that copies are not made or distributed
for profit or commercial advantage and that copies bear this notice and the full citation
on the first page. Copyrights for components of this work owned by others than the
author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or
republish, to post on servers or to redistribute to lists, requires prior specific permission
and/or a fee. Request permissions from [email protected]
KDD’15, August 10-13, 2015, Sydney, NSW, Australia.
Copyright is held by the owner/author(s). Publication rights licensed to ACM.
ACM 978-1-4503-3664-2/15/08 ...$15.00.
DOI: http://dx.doi.org/10.1145 ...
2009 - Connected Action - Marc Smith - Social Media Network AnalysisMarc Smith
Review of social media network analysis of Internet social spaces like twitter, flickr, email, message boards, etc. Network analysis and visualization of social media collections of connections.
This document discusses the design of data for science. It addresses two topics: (1) science through platforms for mediated communication like social media platforms, and (2) science on social science contributions about fundamentals of collaboration. For topic 1, it notes issues with awareness, content sharing, coordination, and longevity on current platforms. For topic 2, it emphasizes the importance of data quality, stability, and social responsibility for science. The document calls for addressing the design of data for science directly through interdisciplinary collaboration between designers, statisticians, computer scientists, data scientists, and social scientists.
First presented at Future of Digital Identity, British Library, 7 Jan 2010. Updated for presentation at Privacy and the Law Conference, London, 1 Dec 2010, and for an OxPILS seminar at Balliol College, 23 May 2012.
This document promotes a social intranet solution from Jive Software. It summarizes that traditional intranets focused only on content without enabling collaboration, but social intranets add social features like profiles, groups, activity streams. This turns employees into information curators and addresses challenges like inefficient searching and information silos. Social intranets drive cross-department collaboration and have proven successful for customers like Alcatel-Lucent, Bupa, and CSC in boosting productivity, engagement, and information sharing. Jive offers an all-in-one social intranet solution with features for mobile access, Outlook integration, and prescriptive services.
The document discusses Members of Congress' use of Twitter to communicate with constituents. It notes that while Members have traditionally used conservative communication methods, Twitter allows for more interactive exchanges. The document analyzes the first year of tweets from Members and finds they mainly use Twitter to distribute vetted content and promote appearances, but not for interactive conversations. It also discusses the potential for Twitter to democratize interactions if used innovatively. Restrictions on Members' communication and the adoption of new technologies are also examined.
This document summarizes a study that compares the use of Facebook Groups and Causes as tools for cyberactivism in Chile and the Concepción area. The study aimed to determine if these applications are used for cyberactivism related to real-world issues and places in Chile. It also sought to identify trends in topics of discussion and whether there are correlations between topics discussed nationally and locally. The main findings were that Facebook is mainly used to strengthen existing social ties rather than create new connections, and that it allows for distributed organization without centralized control of information.
2009-HICSS-42-Best paper-A conceptual and operational definition of ‘social r...Marc Smith
1) The document defines social roles as cultural objects that organize behavior and structure social positions through shared understandings of acceptable actions.
2) It provides a conceptual definition of social roles as patterns of social, structural, and behavioral attributes that can be identified through content analysis and statistical analysis of social networks and behaviors.
3) The paper demonstrates applying this combined conceptual and operational definition to identify key social roles in Usenet and Wikipedia through qualitative and quantitative analysis.
This document discusses the emergence of tools and practices to help people manage the growing amount of information and data. It describes how data visualization tools will play an important role, allowing people to interact with and find patterns in large datasets. These tools will include network diagrams, interactive visualizations that allow user comments and sharing, and visualizations created by foundations to communicate data to broad audiences. The document also notes that social filtering, ambient displays, agents and interfaces will be other important tools to help people cope with information overload in the coming decade.
This presentation was developed to help a client address best practices for building an online community within the workplace. It was based upon a great deal of research and study of the topic and should help those who are seeking information or wish to start an online community, as it pulls together a great amount of data and resources on the topic.
This document outlines a thematic unit on technology and communication that uses the short story "From Email to Bill Gates" by John Seabrook. Over the course of 12 days, students will explore the story, learn about key figures like Bill Gates and Steve Jobs, discuss different forms of social media, and compare/contrast the work and impact of Gates and Jobs. Assessment includes sending an email to Gates, answering text questions, debating topics, and a final exam on the story content and themes. The goal is for students to understand how technology facilitates communication while developing skills like critical thinking.
Crisis Information Management in the Web 3.0 AgeAxel101
The document discusses crisis information management in the Web 3.0 age. It proposes combining social and semantic web data to provide structured and relevant information to enhance situational awareness for emergency responders. A process is described that involves collecting information from citizens and social media, classifying it, enriching it with linked open data, and providing it to decision makers to improve response. Future work could involve user studies and applying machine learning to the classification process.
2010-November-8-NIA - Smart Society and Civic Culture - Marc SmithMarc Smith
This document discusses how social media and social networks are enabling new forms of civic participation and collective action. It notes that citizens are increasingly using social media to find government services, engage in discussions, and measure public opinion. The document also discusses how social network analysis can be used to analyze patterns in social media networks and identify influential users. It provides an overview of various social media platforms and the types of social networks and connections that exist within them.
Social computation of emergent networks on user generated contentMarkus Strohmaier
This document summarizes a presentation on social computational systems given by Markus Strohmaier at the GI Workshop on Web Science in 2010. It discusses the rise of user-generated content and social networks. It also explains that social computational systems emerge from the complex interactions between people and computers on online platforms. Properties of these systems like findability, utility, navigability and relevance can be influenced by social computation processes. The presentation agenda discusses social computational systems, navigability, semantics and the future of these systems. An example is given of how the connectivity of the web is shaped by social mechanisms like preferential attachment.
Digital Trails Dave King 1 5 10 Part 1 D3Dave King
This document provides an overview and agenda for a tutorial on extracting intelligence from digital traces and trails left by web and mobile users. It discusses the proliferation of digital devices that create extensive data about people's online and mobile activities. Examples are given of different types of digital traces, including cookies, web bugs, location data, and social media interactions. Concerns about privacy are also mentioned as vast amounts of personal data are now collected and analyzed.
This document discusses how Twitter can be used to develop a learning community. It provides background on Twitter, noting that it allows rapid dissemination of information to a network of followers. The document then discusses how Twitter can foster informal learning beyond the classroom and help create a community of learners. Some specific ways Twitter has been used for teaching include student notice boards, one-on-one support, summarizing articles, and sharing research links. The document also notes some ethical and legal issues to consider when using Twitter for teaching, such as intellectual property and maintaining student confidentiality. Overall, it argues that when guided properly, Twitter shows potential for enhancing blended learning and developing collaborative learning communities.
This document provides an agenda and background information for a Meet 'N' Greet social event on social computing. The agenda includes a welcome, getting to know each other through brief presentations, and discussing next steps and ongoing activities. The background explains the goals of building an interdisciplinary community and collaboration around social computing research across various University of Minnesota departments and disciplines. Brief biographies of committee members are also included, describing their relevant research interests.
IJERA (International journal of Engineering Research and Applications) is International online, ... peer reviewed journal. For more detail or submit your article, please visit www.ijera.com
Future of the Internet Predictions March 2014 PIP ReportVasily Ryzhonkov
This report is the latest research report in a sustained effort throughout 2014 by the Pew Research Center to mark the 25th anniversary of the creation of the World Wide Web by Sir Tim Berners-Lee. He wrote a paper on March 12, 1989 proposing an “information management” system that became the conceptual and architectural structure for the Web. He eventually released the code for his system — for free — to the world on Christmas Day in 1990. It became a milestone in easing the way for ordinary people to access documents and interact over the Internet — a system that linked computers and that had been around for years.
The Web became a major layer of the Internet. Indeed, for many, it became synonymous with the Internet, even though that is not technically the case. Its birthday offers an occasion to revisit the ways it has made the Internet a part of Americans’ social lives.
Our first report tied to the anniversary looked at the present and the past of the Internet, marking its strikingly fast adoption and assessing its impact on American users’ lives. This report is part of an effort by the Pew Research Center’s Internet Project in association with Elon University’s Imagining the Internet Center to look at the future of the Internet, the Web, and other digital activities. This is the first of eight reports based on a canvassing of hundreds of experts about the future of such things as privacy, cybersecurity, the “Internet of things,” and net neutrality. In this case we asked experts to make their own predictions about the state of digital life by the year 2025. We will also explore some of the economic change driven by the spectacular progress that made digital tools faster and cheaper. And we will report on whether Americans feel the explosion of digital information coursing through their lives has helped them be better informed and make better decisions.
This report is a collaborative effort based on the input and analysis of the following individuals.
THE RELATIONSHIP BETWEEN THE USE OF BLACKBERRY WITH THE STUDENTS’ DEMAND FULF...cscpconf
The communication technology mainly Blackberry enables a medium to facilitate mediated interpersonal communication because of its interactive ability. This aspect creates some
easiness. In the interpersonal communication keeping apart with a distance, the interactivity of the convergent media has been over the potential ability of a feedback since a person accessing
a convergent medium directly gives a feedback of the message conveyed. Blackberry as a result of an advanced technology development has been growing so fast in this life.
Based on the background and phenomenon mentioned, this research studied about “The Relationship between The Use of Blackberry with The Demand Fulfillment and Personality of
The Junior High Students in Bandung.” The research was aimed at finding out: (1) the correlation between ten intensity of blackberry use with the cognitive and affective fulfillment of
The Junior High School Students, (2) the correlation between the intensity of blackberry use with the Tense Release of The Junior High School Students, (3) the correlation between the
intensity of blackberry use with the Personal Integrative of The Junior High School Students; (4) the correlation between the intensity of blackberry use with the Socially Integrative of The
Junior High School Students; (5) the correlation between the intensity of blackberry use with the Confidence of The Junior High School Students; (6) the correlation between the intensity of blackberry use with the Tolerance of The Junior High School Students; (7) the correlation between the intensity of blackberry use with the Whole Fulfillment of The Junior High School
Students; (8) the correlation between the intensity of blackberry use with the Personality as a whole of The Junior High School Students The research used a quantitative approach with the explanatory survey method. The Theories used were: Cognitive Psychology, Technology Determinism, and Uses and Gratification . The population of the research was The Junior High School students. By using random sampling
technique, it was taken 5 schools and 200 students as the sample. The data were taken through questionnaires. The data obtained were analyzes by using the statistical test of correlation. The results of the research were shown in the forms of Bar Chart.
THE RELATIONSHIP BETWEEN THE USE OF BLACKBERRY WITH THE STUDENTS’ DEMAND FULF...csandit
The communication technology mainly Blackberry enables a medium to facilitate mediated
interpersonal communication because of its interactive ability. This aspect creates some
easiness. In the interpersonal communication keeping apart with a distance, the interactivity of
the convergent media has been over the potential ability of a feedback since a person accessing
a convergent medium directly gives a feedback of the message conveyed. Blackberry as a result
of an advanced technology development has been growing so fast in this life.
Based on the background and phenomenon mentioned, this research studied about “The
Relationship between The Use of Blackberry with The Demand Fulfillment and Personality of
The Junior High Students in Bandung.” The research was aimed at finding out: (1) the
correlation between ten intensity of blackberry use with the cognitive and affective fulfillment of
The Junior High School Students, (2) the correlation between the intensity of blackberry use
with the Tense Release of The Junior High School Students, (3) the correlation between the
intensity of blackberry use with the Personal Integrative of The Junior High School Students;
(4) the correlation between the intensity of blackberry use with the Socially Integrative of The
Junior High School Students; (5) the correlation between the intensity of blackberry use with
the Confidence of The Junior High School Students; (6) the correlation between the intensity of
blackberry use with the Tolerance of The Junior High School Students; (7) the correlation
between the intensity of blackberry use with the Whole Fulfillment of The Junior High School
Students; (8) the correlation between the intensity of blackberry use with the Personality as a
whole of The Junior High School Students
The research used a quantitative approach with the explanatory survey method. The Theories
used were: Cognitive Psychology, Technology Determinism, and Uses and Gratification . The
population of the research was The Junior High School students. By using random sampling
technique, it was taken 5 schools and 200 students as the sample. The data were taken through
questionnaires. The data obtained were analyzes by using the statistical test of correlation. The
results of the research were shown in the forms of Bar Chart.
Towards Decision Support and Goal AchievementIdentifying Ac.docxturveycharlyn
Towards Decision Support and Goal Achievement:
Identifying Action-Outcome Relationships From Social
Media
Emre Kıcıman
Microsoft Research
[email protected]
Matthew Richardson
Microsoft Research
[email protected]
ABSTRACT
Every day, people take actions, trying to achieve their per-
sonal, high-order goals. People decide what actions to take
based on their personal experience, knowledge and gut in-
stinct. While this leads to positive outcomes for some peo-
ple, many others do not have the necessary experience, knowl-
edge and instinct to make good decisions. What if, rather
than making decisions based solely on their own personal
experience, people could take advantage of the reported ex-
periences of hundreds of millions of other people?
In this paper, we investigate the feasibility of mining the
relationship between actions and their outcomes from the
aggregated timelines of individuals posting experiential mi-
croblog reports. Our contributions include an architecture
for extracting action-outcome relationships from social me-
dia data, techniques for identifying experiential social media
messages and converting them to event timelines, and an
analysis and evaluation of action-outcome extraction in case
studies.
1. INTRODUCTION
While current structured knowledge bases (e.g., Freebase)
contain a sizeable collection of information about entities,
from celebrities and locations to concepts and common ob-
jects, there is a class of knowledge that has minimal cov-
erage: actions. Simple information about common actions,
such as the effect of eating pasta before running a marathon,
or the consequences of adopting a puppy, are missing. While
some of this information may be found within the free text of
Wikipedia articles, the lack of a structured or semi-structured
representation make it largely unavailable for computational
usage. With computing devices continuing to become more
embedded in our everyday lives, and mediating an increasing
degree of our interactions with both the digital and physical
world, knowledge bases that can enable our computing de-
vices to represent and evaluate actions and their likely out-
comes can help individuals reason about actions and their
Permission to make digital or hard copies of all or part of this work for personal or
classroom use is granted without fee provided that copies are not made or distributed
for profit or commercial advantage and that copies bear this notice and the full citation
on the first page. Copyrights for components of this work owned by others than the
author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or
republish, to post on servers or to redistribute to lists, requires prior specific permission
and/or a fee. Request permissions from [email protected]
KDD’15, August 10-13, 2015, Sydney, NSW, Australia.
Copyright is held by the owner/author(s). Publication rights licensed to ACM.
ACM 978-1-4503-3664-2/15/08 ...$15.00.
DOI: http://dx.doi.org/10.1145 ...
Community Evolution in the Digital Space and Creation of SocialInformation C...Saptarshi Ghosh
A social homogeneous group can be formed irrespective to geo-spatial contiguity and research reveals that interaction through online communication fosters social behaviours like teamwork, ties, bonding and trust building as well as community building.
Selas Turkiye Influence And Passivity In Social Media ExcerptedZiya NISANOGLU
This document discusses influence and passivity in social media. It analyzes how information propagates on Twitter and proposes an algorithm to measure user influence that accounts for both popularity and passivity. The algorithm is evaluated using a dataset of 2.5 million Twitter users and is found to better predict URL clicks and identify influential users compared to other measures like page rank or number of followers. The study finds a weak correlation between popularity and influence, with the most influential users not necessarily the most popular. Highly passive users are more likely to be spammers.
Estudio Influencia Y Pasividad En Social Mediaeliasvillagran
This document discusses influence and passivity in social media. It analyzes how information propagates on Twitter and proposes an algorithm to measure user influence that accounts for both popularity and passivity. The algorithm is evaluated on a dataset of 2.5 million Twitter users and is found to better predict URL clicks and identify influential users compared to other measures like page rank or number of followers. The study finds a weak correlation between popularity and influence, with the most influential users not necessarily the most popular. Highly passive users are more likely to be spammers.
Web Observatories, e-Research and the Importance of Collaboration. WST 2014 Webinar series, 20th March 2014
See Web Science Trust http://webscience.org/
Emerged computer interaction with humanity social computingijcsa
In the 21st century, everywhere people analyze & measure societal. The new trend to compute the societal is
known as social computing. The emerging trend of research focuses interaction of technologies with
humanity. This interaction can be either man machine interaction or human computer interaction. This
article conveys the brief description of social computing and social impact of computing into variant
environment. It optimises the interaction technology, ubiquitous computing and pervasive computing.
Subsequently affective computing is discussed with artificial intelligence to motivate the automation of
technology in social computing.
Sociological Research Methods – Group Research ProjectThe Ev.docxjensgosney
This document provides details for a group sociological research project examining the evolution of technology and its effects on human interactions and communications. It outlines the topic, goals of the study, and methodology. The study aims to discover how interactions between humans have changed with the rise of internet and electronic technology. It will use network ethnography, which allows observation of multiple online networks without limitations of traditional field sites. Follow-up interviews will also be conducted. The document discusses previous related studies and methodology used. It proposes examining how daily smart phone use impacts human relationships through texting, social media, privacy, and time spent on devices. The group will survey 500 smart phone users aged 16-46 and conduct detailed interviews. References several credible sources to
Social networking allows individuals to construct public profiles, connect with others, and view connections within the system. It has evolved from early 20th century ideas to today's major sites like Facebook. Networks can be analyzed at the micro, meso, and macro levels. While social networking provides benefits like strengthening relationships, it also poses privacy and addiction risks. The growth of social networks is evidenced by increasing related patents over time.
Many experts say the rise of embedded and wearable computing will bring the next revolution in digital technology. They say the upsides are enhanced health, convenience, productivity, safety, and more useful information for people/organizations. The downsides: challenges to personal privacy, over-hyped expectations, and boggling tech complexity. Lee Rainie shares the latest research from Pew about libraries and puts it into context with the expanding Internet of Things.
Multimode network based efficient and scalable learning of collective behaviorIAEME Publication
This document discusses multimode network-based approaches for efficiently learning collective behavior in large social networks. It provides an overview of existing approaches for predicting collective behavior based on the behaviors of connected individuals. Specifically, it describes methods that extract social dimensions from networks to represent affiliations between actors and then apply supervised learning to determine which dimensions are informative for behavior prediction. However, existing approaches do not scale well to networks with millions of actors. The document proposes a new edge-centric clustering approach to extract sparse social dimensions, enabling the efficient handling of very large networks while maintaining predictive performance.
303 final project-Communication Technology in the year 2024jonmitch502
The document discusses communication technologies and how they will evolve by 2024. It outlines several theories about the diffusion of innovations and adoption of new technologies. It predicts that by 2024, computers and internet access will be ubiquitous and integral to work and education. Social media sites will be more advanced and integrated into other services. E-readers and online learning will likely replace physical books and traditional classroom-based education. Communication technologies will continue to rapidly develop and change the ways people interact and obtain information.
2014 IE Application- Social Interaction SJWSeebas85
This document discusses how technology will impact social interaction over the next 10 years. It outlines that technology has historically facilitated social interaction through innovations like writing, printing, and telecommunications. Going forward, anticipated innovations like augmented reality and smart clothing could reshape socializing. However, technology may also lead to more superficial "broad and shallow" relationships as people have more connections but spend less face-to-face time together. Additionally, individuals and companies may commercialize personal data, and people may gain more awareness of how technology shapes their interactions and take steps to limit overdependence on digital connections. Ultimately, how technology affects well-being will depend on how it is used within society.
The document is a research proposal examining the impact of social networking sites on the academic performance of adolescents. It includes an introduction outlining the research problem, literature review on relevant topics, research design using questionnaires and interviews of a sample of local adolescents, and a proposed presentation of findings and data analysis. The research aims to evaluate the negative impact of social media usage on academics, determine influencing factors for student participation, and identify measures to reduce time spent on social networking sites.
Using ICTs to Promote Cultural Change: A Study from a Higher Education Contextac2182
The document summarizes Anthony Cocciolo's research on how the introduction of a Web 2.0 technology impacts the culture of learning within a community. The research studied a system called PocketKnowledge over two years. Results showed evidence that the technology provided a space for a participatory subculture to form, as users were more likely to both consume and contribute content than on other sites. This visibility of contributions online may impact users' decisions to participate and the overall culture. Future research could explore how to specifically design technologies to support participatory cultures and cultural change.
This study examines a phenomenon called the "informatics moment" that occurs when library patrons seek help using new digital technologies, such as computers or the internet, at public libraries. The study identifies four types of literacy involved in these moments: basic literacy, computer literacy, library literacy, and domain literacy, which is most commonly related to job searches. These informatics moments provide opportunities to understand how social capital, such as help from familiar people, can help individuals and societies transition to the digital age by overcoming barriers. The study aims to understand both the informatics moment itself and how social capital affects the process and outcomes of these help-seeking interactions.
This document discusses a dissertation submitted by Abah Stephany Mbong to the University of Bamenda in partial fulfillment of the requirements for a Higher Technical Teacher School Post Graduate Diploma in Information Management and Communication. The dissertation examines the effects of social networking services on the output of an organization. It includes a title page, certification, attestation, dedication, acknowledgements and table of contents. The literature review discusses concepts of social networking services, their impact and features. Examples of social networking sites discussed include 2go, WhatsApp Messenger, Blackberry Messenger, Facebook and Twitter.
Suazo%2c martínez & elgueta english version2011990
This document summarizes a study that analyzed the use of Facebook Groups and Causes as tools for cyberactivism in Chile and the Concepción area. The study found that social networks mainly strengthen existing social ties rather than create new connections. It also found that Facebook's horizontal structure and lack of centralized control enables strong information sharing. Finally, the study aimed to determine if Groups or Causes were better for generating cyberactivism campaigns and setting independent social media agendas.
Similar to Validation of Dunbar's number in Twitter conversations (20)
Franco, Augusto (2017) Conservadorismo, liberalismo econômico e democraciaaugustodefranco .
O documento resume dez princípios conservadores propostos por Russell Kirk. O primeiro princípio é a crença em uma ordem moral duradoura, baseada na ideia de que a natureza humana é constante e as verdades morais são permanentes. Entretanto, essa visão de ordem moral pré-existente pode levar a uma justificação autocrática do poder em vez de democrática.
Franco, Augusto (2018) Os diferentes adversários da democracia no brasilaugustodefranco .
O documento discute a classificação de forças políticas proposta por Takis Pappas para a Europa e argumenta que ela não se aplica ao Brasil. Propõe uma nova classificação com antidemocratas (de direita e esquerda), neopopulistas e populistas autoritários como adversários da democracia no país.
1) A democracia está sob ataque em todo o mundo e o número de democracias não tem aumentado no século 21. 2) Populistas e neopopulistas autoritários como Trump, Putin e outros estão ganhando poder e ameaçando a democracia. 3) Esquerda e direita se unem contra a globalização e a favor do autoritarismo, representando uma ameaça à democracia e à emergência de uma sociedade em rede.
El documento presenta la idea de que existe una "Matrix realmente existente" que deforma el campo social y hace que las personas se comporten de manera replicante. La Matrix se carga en las personas a través de la familia, la escuela, la iglesia, las organizaciones y el trabajo, haciéndolas actuar como si estuvieran bajo la influencia de un sistema de creencias y valores compartidos. Sin embargo, las personas no se dan cuenta de que al actuar basadas en estas creencias básicas, están reproduciendo una realidad social deformada.
Algumas notas sobre os desafios de empreender em redeaugustodefranco .
O autor descreve sua tentativa de configurar um "cluster de inteligência colaborativa" com amigos para empreender em rede. No entanto, ele observa que é mais fácil projetar do que implementar esse tipo de empreendimento distribuído, devido a fatores ambientais e estruturais que dificultam a emergência da inteligência coletiva. Ele argumenta que ambientes hierárquicos impedem o tempo necessário de interação para que esses sistemas se desenvolvam de forma autônoma e estável.
1) Cortar custos é necessário durante crises, mas as empresas devem cortar os "custos invisíveis", como custos de transação, sinergia e atrito de gestão, ao invés de demitir funcionários. 2) Esses custos invisíveis, como burocracia e falta de colaboração, podem ser maiores do que os custos aparentes. 3) Transformando a organização e gestão da empresa em sistemas mais baseados em rede pode reduzir esses custos e trazer mais inovação e sustentabilidade à longo prazo.
1) O documento discute duas perspectivas para explicar o fenômeno da aprendizagem: a perspectiva informacionista e a perspectiva da deriva ontogênica.
2) O autor argumenta que a perspectiva informacionista requer interações instrutivas e não é compatível com sistemas determinados estruturalmente como os seres vivos.
3) A perspectiva da deriva ontogênica é compatível com sistemas determinados estruturalmente e explica a aprendizagem como resultado da história de interações e mudanças estruturais congr
CONDORCET, Marquês de (1792). Relatório de projeto de decreto sobre a organiz...augustodefranco .
1) A instrução pública deve ser universal, igualitária e abranger todas as idades para garantir que todos os cidadãos tenham acesso ao conhecimento.
2) Os estabelecimentos de ensino devem ser independentes de qualquer autoridade política para assegurar a livre disseminação do conhecimento.
3) A liberdade do ensino é essencial para o progresso das ciências e das luzes, e nenhum poder deve ter o direito de limitar o ensino de novas ideias.
NIETZSCHE, Friederich (1888). Os "melhoradores" da humanidade, Parte 2 e O qu...augustodefranco .
Nietzsche critica os "melhoradores" da humanidade que tentam seduzir as pessoas para dentro de instituições como a Igreja. Ele argumenta que a Igreja tornou as pessoas fracas ao invés de melhorá-las e que coisas boas como a beleza raramente podem ser compartilhadas por todos. Nietzsche também afirma que os alemães carecem de espírito e coragem para defender suas próprias opiniões.
1) O documento descreve um projeto de investigação-aprendizagem sobre padrões autocráticos em obras de ficção e filmes.
2) O objetivo é identificar padrões de organização e regulação hierárquica presentes em diferentes contextos, como no nazismo, marxismo e ditaduras.
3) Um exemplo dado é o culto ao trabalho presente nos campos de concentração nazistas e soviéticos.
O documento discute as principais ameaças à liberdade e paz mundial: 1) o neo-expansionismo russo sob Putin e 2) a expansão do jihadismo islâmico. Não há solução fácil para essas ameaças estatistas senão uma "mudança de época" através de uma democratização radical das sociedades que deslegitime os Estados e proto-Estados autoritários.
Textos de Augusto de Franco publicados no Facebook, entre maio e setembro de 2014, sobre o Decreto 8.243/2014 que institui a Política Nacional de Participação Social - PNPS e o Sistema Nacional de Participação Social - SNPS, e dá outras providências
1) O autor argumenta que as eleições nem sempre consolidam a democracia e podem, na verdade, conspirar contra o processo de democratização em países com governos autocráticos.
2) Ele aponta exemplos recentes como Síria, Egito e Turquia onde eleições foram realizadas, mas não promoveram a democracia.
3) No Brasil, o autor afirma que o processo eleitoral vem sendo usado pelo governo do PT para enfraquecer a democracia, ao invés de consolidá-la.
1. O documento descreve um programa de aprendizagem democrática baseado na leitura de romances e filmes clássicos que abordam temas como distopias e autocracias.
2. O programa consiste em 10 módulos realizados ao longo de 100 dias, cada um contendo um romance e um filme, além de um teste.
3. Os fundamentos do programa incluem o estudo do "padrão Darth Vader" de comportamento autoritário para reconhecê-lo onde quer que surja.
Democracia cooperativa: escritos políticos escolhidos de John Deweyaugustodefranco .
FRANCO, Augusto e POGREBINSCHI, Thamy (editores) (2008) Democracia Cooperativa - Escritos Políticos Escolhidos de John Dewey. A versão final desta obra foi impressa em papel. Porto Alegre: ediPUCRS, 2008.
Como configurar um ambiente de investigação-aprendizagem openscience em sua escola. Capacitação teórica e opção de programa prático para educadores e gestores escolares e extraescolares.Programa intensivo de aprendizagem para agentes de educação que desejam configurar novos ambientes inovadores de pesquisa ou investigação em escolas e outras organizações. Programa presencial ofertado pelo LABE=R em São Paulo.
Programa de Aprendizagem Política a distância, ofertado pelo LABE=R (Laboratório da Escola-de-Redes), com 80 horas (estimadas) de duração, realizado totalmente pelo Facebook, de junho a setembro de 2014.
This document from Human Rights Watch investigates alleged human rights violations committed by Venezuelan security forces in response to protests since February 2014. It documents 45 cases of serious abuses, including unlawful killings, torture, arbitrary arrest and detention, and violations of due process rights. The abuses appeared to target both protesters and those documenting the security response, and occurred across multiple states. Prosecutors and judges failed to properly investigate and enable accountability. While some protesters have used violence, evidence suggests many victims of security force abuses were nonviolent or not engaged in criminal activity at the time.
Validation of Dunbar's number in Twitter conversations
1. Validation of Dunbar’s number in Twitter conversations
Bruno Gon¸alves1,2 , Nicola Perra1,3 , and Alessandro Vespignani1,2,4
c
1
Center for Complex Networks and Systems Research,
School of Informatics and Computing, Indiana University, IN 47408, USA
2
Pervasive Technology Institute, Indiana University, IN 47404, USA
3
Linkalab, Complex Systems Computational Lab. - 09100 Cagliari Italy and
4
Institute for Scientific Interchange, Turin 10133, Italy
Modern society’s increasing dependency on online tools for both work and recreation opens up
unique opportunities for the study of social interactions. A large survey of online exchanges or
conversations on Twitter, collected across six months involving 1.7 million individuals is presented
here. We test the theoretical cognitive limit on the number of stable social relationships known as
arXiv:1105.5170v1 [physics.soc-ph] 25 May 2011
Dunbar’s number. We find that users can entertain a maximum of 100 − 200 stable relationships
in support for Dunbar’s prediction. The “economy of attention” is limited in the online world by
cognitive and biological constraints as predicted by Dunbar’s theory. Inspired by this empirical
evidence we propose a simple dynamical mechanism, based on finite priority queuing and time
resources, that reproduces the observed social behavior.
I. INTRODUCTION measure typical group size using two different methods
and obtained a number of 291, roughly twice Dunbar’s
Modern society’s increasing dependence on online estimate.
tools for both work and recreation has generated an Biological constraints on social interaction go along with
unprecedented amount of data regarding social behavior. other real-world physical limitations. After all, a persons
While this dependence has made it possible to redefine time is finite and each person must make her own choices
the way we study social behavior, new online commu- about how best to use it given the priority of personal
nication tools and media are also constantly redefining preferences, interests, needs, etc. The idea that attention
social acts and relations. Recently, the divide between and time are scarce resources led H. Simon [10] to apply
the physical world and online social realities has been standard economic tools to study these constraints
blurred by the new possibilities afforded by real-time and introduce the concept of an Attention Economy
communication and broadcasting, which appear to with mechanisms similar to our everyday monetary
greatly enhance our social and cognitive capabilities in economy. The increasingly fast pace of modern life and
establishing and maintaining social relations. The com- overwhelming availability of information has brought
bination of mobile devices with new tools like Twitter, a renewed interest in the study of the economy of
Foursquare, Blippy, Tumblr, Yahoo! Meme, Google attention with important applications both in business
Hotspot, etc., are defining a new era in which we can be [11] and the study collective human behavior [12]. On
continuously connected with an ever-increasing number the one hand it can be argued that microblogging tools
of individuals through constant digital communication facilitate the way we handle social interactions and
composed of small messages and bits of information. that this results in an online world where human social
Thus, while new data and computational approaches to limits are finally lifted, making predictions such as the
social science [1–3] finally enable us to answer a large Dunbar’s number obsolete. Microblogging and online
number of long-standing questions [4–6], we are also tools on the other hand, might be analogous to a pocket
increasingly confronted with new questions related to calculator that, while speeding up the way we can do
the way social interaction and communication change simple math, does not improve our cognitive capabilities
in online social environments: What is the impact that for mathematics. In this case, the basic cognitive
modern technology has on social interaction? How do limits to social interactions are not surpassed in the
we manage the ever-increasing amount of information digital world. In this paper we show that the latter
that demands our attention? In 1992, R. I. M. Dunbar hypothesis is supported by the analysis of real world
[7] measured the correlation between neocortical volume data that identify the presence of Dunbars limit in Twit-
and typical social group size in a wide range of primates ter, one of the most successful online microblogging tools.
and human communities. The result was as surprising
as it was far-reaching. The limit imposed by neocortical
processing capacity appears to define the number of
individuals with whom it is possible to maintain stable II. THE DATASET
interpersonal relationships. Therefore, the size of the
brain’s neocortex represents a biological constraint on Having been granted temporary access to Twitters fire-
social interaction that limits humans’ social network size hose we mined the stream for over 6 months to identify a
to between 100 and 200 individuals [8], i.e. Dunbar’s large sample of active user accounts. Using the API, we
number. McCarty et al. [9] independently attempted to then queried for the complete history of 3 million users,
2. 2
A) B) k in = 3 k in = 1
k out = 3 k out = 1
w in = 4 w in = 1
4 w out = 3 w out = 2
Alice
10 B C
Ellie k in = 2
2 k out = 1
Bob w in = 2
7 w out = 1
Bob
1 5 A
Alice Cathy
9
E
Cathy
k in = 1 k in = 0
3 k out = 1 k out = 1
Dan w in = 1 w in = 0
11 D w out = 1 w out = 1
Bob
6
Bob
FIG. 1: Reply trees and user network. A) The set of all trees is a forest. Each time a user replies, the corresponding tweet
is connected to another one, resulting in a tree structure. B) Combining all the trees in the forest and projecting them onto
the users results in a directed and weighted network that can be used as a proxy for relationships between users. The number
of outgoing (incoming) connections of a given user is called the out (in) degree and is represented by kout (kin ). The number
of messages flowing along each edge is called the degree, ω. The probability density function P (kout ) (P (kin )) indicates the
probability that any given node has kout (kin ) out (in) degree and it is called the out (in) degree distribution and is a measure
of node diversity on the network.
resulting in a total of over 380 million individual tweets about the id of the original tweet but also the user that
covering almost 4 years of user activity on Twitter. Ta- sent it. Using this information, each reply tweet maps
ble I provides some basic statistics about our dataset. directly to a directed edge. Individual trees can be iden-
Here we analyze this massive dataset of Twitter conver- tified by using depth first search [17] to identify connected
sations accrued over the span of six months and investi- components in the resulting tweet-tweet graph. To en-
gate the possibility of deviation from Dunbar’s number sure that the full tree is found and not just a part of
in the number of stable social relations mediated by this it, we treat each link as undirected for the purposes of
tool. The pervasive nature of Twitter, along with its this identification. In this way we are able to extract the
widespread adoption by all layers of society, makes it an complete tree even if we happen to start on one of the
ideal proxy for the study of social interactions [13–16]. leaves. For each tree the root is then found by locating
We have analyzed over 380 million tweets from which the node with kin ≡ 0, and distances from the root are
we were able to extract 25 million conversations. Each measured by rerunning the DFS algorithm starting from
Twitter conversation takes on the form of a tree of tweets, the root and respecting the direction of each edge.
where each tweet comes as a reply to another. By pro- The underlying reply network can be extracted by pro-
jecting this forest of trees onto the users that author each jecting the tweet trees to a user graph: User A is con-
tweet, we are able to generate a weighted social network nected to user B by a directed outgoing edge if A replied
connecting over 1.7 million individuals (see Figure 1). to a tweet sent by B. Over time, any pair of users can ex-
change multiple replies either in a single “conversation”
Tweets 381, 652, 990 (tree) or through multiple conversations. The number of
Timelined Users 3, 006, 180 messages sent from one user to another is used as the
Scraping Period Nov. 20, 2008 – May 29, 2009
weight of the corresponding directed edge and is taken
to signify the strength of the connection between the two
Time span 4 years
users, with higher weights representing stronger connec-
tions.
Trees 25, 273, 871
Tweets in Trees 81, 728, 252
User in Trees 1, 720, 320 B. Online conversations
User-User Edges 68, 459, 592
Each reply creates a connection between two tweets
TABLE I: Dataset Statistics. and their authors, so we can define a conversation as
a branching process of consecutive replies, resulting in
a tree of tweets. From our dataset we extracted and
analyzed a forest of over 25 million trees. Trees vary
A. Tree Identification and Projection broadly in size and shape, with most conversations re-
maining small while a few grow to include thousands of
All tweets in our dataset that constituted a reply were tweets and hundreds of users, as shown in Figure 2.
collected. Each such tweet contains information not only A directed user-user network can be built by projecting
3. 3
0 0
10 10
A) B)
slope = -3 slope = -5
-2
10 10
-2
-4
10 -4
10
P(Ns)
P(Ds)
-6
10 -6
10
-8
10 -8
10
-10
10
-10
10
-12
10 0 1 2 3 4 5 0 1 2 3 4
10 10 10 10 10 10 10 10 10 10 10
Ns Ds
0 4
10 10
C) slope = 1
D)
slope = -3
-2
10
3
10
-4
10
P(Nu)
-6
Nt 10
2
10
-8
10 1
10
-10
10
0
0 1 2 3 4 10 0 1 2 3 4
10 10 10 10 10 10 10 10 10 10
Nu Ds
FIG. 2: Tree characterization. A)Distribution of the number of tweets in a tree. B) Distribution of the number of shells. C)
Distribution of the number of users. D) Tree size vs depth. The broad tailed nature of all of these quantities indicates the
diversity of behaviors displayed by the users in our dataset.
conversation trees to detail how users interact and estab- to the number of different nodes it receives a reply from.
lish relationships among themselves. Bidirectional edges When A follows B, A subscribes to receive all the up-
signify mutual interactions, with stronger weights imply- dates published by B. A is then one of B’s followers and
ing a more frequent or prolonged interaction between two B is one of A’s friends. Previous studies have mostly
individuals. focused on the network induced by this follower-friend
All of our analysis will be performed on this user-user relationship [15], [18–20]. In any study about stable so-
conversation network. We consider a user to have out cial relations in online media, as indicated by studies
degree kout if he or she replies to kout other users, re- about Dunbar’s number, it is important to discount oc-
gardless of the number of explicit followers or friends the casional social interactions. For this reason we focus on
given user has. By focusing on direct interactions we are stronger relationships in our study[13], considering just
able to eliminate the confounding effect of users that have active communication from one user to another by means
tens or hundreds of thousands of followers with whom of a genuine social interaction between them. In our net-
they have no contact and are able to focus on real person work [21, 22] we introduce the weight ωij of each edge,
to person interactions [13]. defined as the number of times user i replies to user j as
a direct measurement of the interaction strength between
two users and stable relations will be those with a large
III. DUNBAR’S NUMBER IN OUR DATA weight. A simple way to measure this effect is to calculate
the average weight of each interaction by a user as a func-
In the generated network each node corresponds to a tion of his total number of interactions. Users that have
single user. The out-degree of the nodes is the number of only recently joined Twitter will have few friends and
users the node replies to, while the in-degree corresponds very few interactions with them. As time goes by, stable
4. 4
users will acquire more and more friends, but the number are the keys ingredients in fixing the Dunbar’s number.
of replies that they send to other users will increase con- We model this considering that when user i receives a
sistently only in stable social interactions. Eventually, a message it places it in an internal queue that allows up
point is reached where the number of contacts surpasses to qmax,i messages to be handled at each time step. In
the user’s ability to keep in contact with them. the presence of finite resources each agent has to make
This saturation process will necessarily lead to some rela- decisions on what are the most important messages to
tionships being more valued than others. Each individual answer. We set the priority of each message to be pro-
tries to optimize her resources by prioritizing these inter- portional to the total degree of the sender j. For each
actions. To quantify the strength of these interactions, user the we studied is the average number of interactions
out out
we studied the quantity ωi , defined as the average so- per connection ωi (T ) as defined in the Eq. (1). At each
cial strength of active initiate relationship: time step each agent goes through its queue and performs
the following simple operations:
out j ωij (T )
ωi (T ) = out (1)
ki • The agent replies to a random number St of mes-
sages between 0 and the number of messages qi
This quantity corresponds to the average weight per present in the queue. The messages to be replied
outgoing edge of each individual where T represents the to are selected proportionally to the priority of the
time window for data aggregation. We measure this sending agent (its total degree). A message is then
quantity in our data set as shown in Figure 3A. The data sent to j, the node we are replying to, and the cor-
shows that this quantity reaches a maximum between 100 responding weight ωij is incremented by one.
and 200 friends, in agreement with Dunbar’s prediction
(see figure 2A). This finding suggests that even though
• Messages the agent has replied to are deleted from
modern social networks help us to log all the people with
the queue and all incoming messages are added to
whom we meet and interact, they are unable to overcome
the queue in a prioritized order until the number of
the biological and physical constraints that limit stable
messages reaches qmax . Messages in excess of qmax
social relations. In Figure 2B, we plot , the number of
are discarded.
reciprocated connections, as a function of the number
of the in-degree. saturates between 200 and 300 even
though the number of incoming connections continues to The dynamic process is then repeated for a total num-
increase. This saturation indicates that after this point ber of time steps T . In order to initialize the process
the system is in a new regime; new connections can be re- and take into account the effect of endogenous random
ciprocated, but at a much smaller rate than before. This effects, each agent can broadcast a message to all of its
can be accounted for by spurious exchanges we make with contacts with some small probability p. One may think of
some contacts with whom we do not maintain an active this message as a common status change, or a TV appear-
relationship. ance, news story, or any other information not necessarily
authored by the sending agent. Since these messages are
not specifically directed from one user to another, they do
IV. THE MODEL not contribute to the weight of the edges through which
they flow. We have studied this simple model by using an
underlying network of N = 105 nodes and different scale-
Let us consider a static network G, characterized by a
free topologies. For each simulation T = 2 × 104 time
degree distribution P (k). Each user (node) i is connected
steps have been considered and the plots are made evalu-
to all its nearest neighbors js through two weighted di-
ating the medians among at least 103 runs. In Figure 4 we
rected edges, i → j and j → i so that:
report the results of simulations in a directed heavy-tailed
out in ki network with a power-law tail similar to those observed
ki = ki = , ∀i ∈ G. (2) for the measured network [19]. The figures clearly show a
2
behavior compatible with the empirical data. The peak
out
Where ki is the out degree, the number of out-going that maximizes the information output per connection
in
links, and ki is the in degree, the number of in-going is linearly proportional to qmax , supporting the idea that
links, of the user i. Each node uses its out links to the physical constraints entailed in the queue’s maximum
send messages to its contacts and it will receive messages capacity along with the prioritization that gives impor-
from its contacts through its in links. In this way is easy tance to popular senders are at the origin of the observed
to distinguish between incoming and outgoing messages. behavior. We have also performed an extensive sensitiv-
Whenever a message is sent from node i to node j, the ity analysis on the broadcasting probability p, the time
weight of the (i, j) edge, wij is increased by one. The to- scale T , and have investigated the effect of agent hetero-
tal number of sent messages of each user is given by the geneity by studying populations in where each agent’s
sum over all of its outgoing edges. Users communicate capacity qmax ,i is randomly distributed according to a
with each other by replying to messages. The assump- Gaussian distribution centered around qmax with stan-
tion of our model is that biological and time constraints dard deviation σ.
5. 5
A) B)
50 100 150 200 250 300 350 400 450 500 550 600
8
7
6
5
ωout
ρ
4
3
2
0
1
0 50 100 150 200 250 300 350 400 450 500 550 600 0 50 100 150 200 250 300 350 400 450 500 550 600
out in
k k
FIG. 3: A) Out-weight as a function of the out-degree. The average weight of each outward connection gradually increases
until it reaches a maximum near 150 − 200 contacts, signaling that a maximum level of social activity has been reached. Above
this point, an increase in the number of contacts can no longer be sustained with the same amount of dedication to each. The
red line corresponds to the average out-weight, while the gray shaded area illustrates the 50% confidence interval. B) Number of
reciprocated connections, ρ, as a function of kin . As the number of people demanding our attention increases, it will eventually
saturate our ability to reply leading to the flat behavior displayed in the dashed region.
queues start to get messages in and out. After a while
3
20
x 10 we can aspect that the system reaches a dynamical equi-
80
slope = 0.2
qmax = 50 librium. In Figure (5-A) we show the behavior of our
60
qmax = 100 observable ω out for different values of T , in particular we
Peak
40
15 qmax = 150
20
qmax = 200 chose T = 104 , 1.5 × 104 , 2 × 104 . The effect of time is
0
qmax = 250
0 50 100 150 200 250 300
qmax
qmax = 300
clearly a shift on the y axis and a small change in the
out
10 position of the peak. The first effect is due to the fact
ω
that the number of messages circulating in the systems
5 increase linearly with T . The second effect is due to the
reduction of fluctuations when more messages are sent.
0
The peak becomes more clear and defined.
0 50 100 150 200 250
out
k
B. Effect of broadcast probability p
FIG. 4: Result of running our model on a heterogeneous
network made of N = 105 , nodes with degree distribution
P (k) k−γ with γ = −2.4 and σ = 10. Different curves cor- The effect of the broadcast probability is different on
respond to different queue size. The inset shows the linear respect to the effect of the time window T . First of all
dependence of the peak on the queue size q. Each curve is our observable ω out is linearly proportional to T in all
the median of 103 to 2 × 103 runs of T = 2 × 104 time steps regimes of k out this is not true for p. The effect of p is
crucial for users with a small number of contacts. As the
p increases they will receive more messages and their ac-
tivity will increase too, this does not occur in the other
A. Effect of the time window T
limit. When the saturation takes place the ω out becomes
completely independent of p. As show in details in a
One of the parameters of our model is the time window mean-field approach (Section (IV D)) for values of k out
T during which we study the dynamics. This parame- small with respect to the queue size, ω out scales linearly
ter regulates the maximum number of messages that will with p. Instead for a number of contacts much bigger
circulate in the network. In the first time steps the first than the queue size ω out is independent of p. These con-
messages will start to being sent among users and the siderations are validated by our simulations as shown in
6. 6
Figure (5-B). We see a clear dependence on p for small as the number of in-coming connections, the number of
values of k out instead the same behavior for bigger values messages it get scale as the square of its degree.
of k out . Two different regimes are easily found: ki qmax,i and
vice versa.
In the first case the user is not popular. The number
C. Effect of network’s properties of messages that the user will receive is small then. In
principle it can reply to all of them at each time step.
Inspired by several studies [13, 15, 19] we fix the base- We can assume that in this regime its queue is never
line of our model using scale-free networks. It is im- completely full. We will refer to Rt as the number of
portant then to study how differences in the network messages that the user reiceive at the time step t. After
structure affect the results. In this section we consider one time step the number of replies is:
the effect of the exponent γ. As show in Figure (5-C)
we run our model on top of scale-free networks with S1 = ξ1 R1 , (6)
γ = −2.2, −2.4, −2.6, −2.8. As clear from the plot for
smaller values of γ (bigger value in absolute value) gaps where ξ1 is a random number uniformly distributed be-
on k out start to emerge. These are due to the network tween 0 and 1. The number of messages, S2 that the user
structure. The shape and position of the peak is the send at the second time step is a random fraction of the
same for all the curves. The differences are evident just messages present in its queue:
on the peak height that increase as γ decreases. This is
S2 = [R1 (1 − ξ1 ) + R2 ] ξ2 . (7)
due the different redistribution of degrees and to the fact
that with small γ the selection effect is more and more For t = 3 we get:
important. So we can say that the result are robust on
γ. S3 = {[R1 (1 − ξ1 ) + R2 ] (1 − ξ2 ) + R3 } ξ3
= [R1 (1 − ξ1 )(1 − ξ2 ) + R2 (1 − ξ2 ) + R3 ] ξ3 , (8)
D. Single user: analytical approach
and so on. We can approximate these equations using
the average number of received messages R . For the
In order to get a better understanding of the mecha- general t it is possible to show that:
nisms we describe, we analyzed, in a mean-field approach,
the behavior of a single user i. t t−1
Let us focus on a user i characterized by degree ki and St ∼ R ξt 1 + (1 − ξi )
out
qmax,i . ki = ki /2 are the out-going links that it uses j=1 i=j
in
to send messages to its ki /2 contacts. ki = ki /2 are
the in-coming through which it receives messages from = R ξt 2 − ξt−1 + O(ξ 2 )
it contacts. We set as kj the priority of each neighbor = R 2ξt − ξt ξt−1 + O(ξ 3 ) . (9)
that we extract for a distribution P(k). The rules of the
model that we described in the previous sections are ap- The total number of messages sent is the numerator of
plied for T time steps. The probability that a neighbor our measure ω out and the sum of all the St :
j will send a message to the user i is: t=T
ki St ∼ T R , (10)
pji = p + , (3) t=0
< k > kj
where p is the broadcast probability. We can evaluate the considering that each sum of product random numbers is
average number of messages that the user will receive at order T . We can write then:
each time step t: t=T out
t=0 St 1 ki (k out )2
out
ωi (T ) = out ∼ out T p +c i
in ki 1 ki ki 2 2<k>
R = pji = pki + . (4)
j
<k> in
kj ∼ T out
ki . (11)
j∈ki
We extracted kj from the same distribution, the sum
out
in
scales then linearly with the number of element: ki . We In this regime we get a linear increase with ki of the
can write: average number of replies per connections. As show in
2
Figure (6) this is confirmed in the simulations.
ki ki The other regime is found for a number of contacts bigger
R = pji = p +c , (5)
j
2 2<k> than the queue size. In this case the user is very popular
and at each time step it gets a lot of messages and is not
where c is a constant fixed by the distribution. Since the able to handle all of it. In this limit the saturation process
priority of the user is proportional to its degree as well takes place and it will reply just to a small fraction of
7. 7
3 3 3
x 10 x 10 x 10
A) 10
T = 10
4 B) 20
p = 10
-4
C) γ = -2.2
8 4 -4 γ = -2.4
T = 1.5 x 10 p = 5x10 10
4 15 γ = -2.6
T = 2 x 10 p = 10
-3
γ = -2.8
6
out
out
out
10
ω
ω
ω
4 5
5
2
0 0 0
0 50 100 150 200 250 0 50 100 150 200 250 0 50 100 150 200 250 300
out out out
k k k
FIG. 5: A) ω out as a function of kout for qmax = 100, σ = 10, scale-free network with γ = −2.4 and different values of T . We
present the medians over 500 runs B) ω out as a function of kout for qmax = 100, σ = 10, scale-free network with γ = −2.4,
T = 104 and different values of p. We present the medians over 500 runs C) ω out as a function of kout for qmax = 100, σ = 10,
T = 104 , p = 5 × 10−4 , scale-free network with different values of γ. We present the median over 500 runs.
In this regime then we get a different scaling behavior
x 10
3 typical of saturation problems. As shown in Figure (6)
1
these arguments are in perfect agreement with the nu-
σ = 20
σ = 10 merical results.
σ=5
slope = +1 We have shown two different regimes. A linear increasing
slope = -1
behavior and a decreasing one. In the between of these
opposite cases we will find a maximum of the function.
out
The position of these peak is in general function of the
ω
queue size.
0.1
1 10 100 1000
out
k
FIG. 6: Results for the single user and different values of σ, V. CONCLUSIONS
the inter-user queue size variance. We fixed the average queue
size at qmax,i = 50 and extracted the priorities of user neigh-
bors from a power-law statistical distribution with exponent
γ = −2.1. For each ki we run T = 500 time steps and present Social networks have changed they way we use to com-
the medians among 103 runs municate. It is now easy to be connected with a huge
number of other individuals. In this paper we show that
social networks did not change human social capabili-
the total number of messages prioritizing them. At each ties. We analyze a large dataset of Twitter conversations
time step this number is a random variable uniformly collected across six months involving millions of individ-
distributed between 0 and qmax,i . We have then: uals to test the theoretical cognitive limit on the number
of stable social relationships known as Dunbar’s num-
T ber. We found that even in the online world cognitive
out 1
ωi (T ) ∼ out ξt qmax,i . (12) and biological constraints holds as predicted by Dunbar’s
ki t=0 theory limiting users social activities. We propose a sim-
ple model for users’ behavior that includes finite priority
The ξt s are random variable uniformly distributed be- queuing and time resources that reproduces the observed
tween 0 and 1. At each time step the number of replies is social behavior. This simple model offers a basic explana-
a random fraction of the queue size. For T large enough tion of a seemingly complex phenomena observed in the
we get: empirical patterns on Twitter data and offers support to
T Dunbar’s hypothesis of a biological limit to the number
out
ωi (T ) ∼ out qmax,i . (13) of relationships.
2ki
[1] David Lazer, Alex Pentland, Lada Adamic, Sinan Aral, takis, Noshir Contractor, James Fowler, Myron Gut-
Albert-Laszlo Barabasi, Devon Brewer, Nicholas Chris-
8. 8
mann, Tony Jebara, Gary King, Michael Macy, Deb Roy, 2007.
and Marshall Van Alstyne. Computational social science. [13] B. A. Huberman, D. M. Romero, and F. Wu. Social net-
Science, 323:721, Feb 2009. works that matter: Twitter under the microscope. First
[2] D. J. Watts. The new science of networks. Ann. Rev. Monday, 14:1, 2008.
Socio., 30:243, 2004. [14] A. Lashinsky. Discover. decipher. disrupt. the true mean-
[3] Adrian Cho. Ourselves and our interactions: The ulti- ing of twitter. Fortune Magazine, 158:39, 2008.
mate physics problem? Science, 325:406, Nov 2009. [15] D. Boyd, S. Golder, and G. Lotan. Tweet, tweet, retweet:
[4] A.-L. Barab´si. The origin of bursts and heavy tails in
a Conversational aspects of retweeting on twitter. In 43rd
human dynamics. Nature, 435:207, 2005. Hawaii International Conference on System Sciences,
[5] C. Castellano, S. Fortunato, and V. Loreto. Statisti- page 412, 2008.
cal physics of social dynamics. Rev. Mod. Phys., 81:591, [16] Haewoon Kwak, Changhyun Lee, Hosung Park, and Sue
2009. Moon. What is twitter, a social network or a news me-
[6] M. C. Gonz´lez, C. A. Hidalgo, and A.-L. Barab´si. Un-
a a dia?, 2010.
derstanding individual human mobility patterns. Nature, [17] T. H. Cormen, C. E. Leiserson, R. L. Rivest, and C. Stein.
453:779, 2008. Introduction To Algorithms. The MIT Press, 2nd edition,
[7] R. I. M. Dunbar. Neocortex size as a constraint on group 2001.
size in primates. J. Human Evo., 22:469, 1992. [18] B. Krishnamurthy, P. Gill, and M. Arlitt. A few chirps
[8] R. I. M Dunbar. The social brain hypothesis. Evo. An- about twitter. In Proceedings of the first workshop on
thro., 6:178, 1998. Online social networks, page 19, 2008.
[9] Christopher McCarty, Peter D Killworth, H Russell [19] A. Java, X. Song, T. Finin, and B. Tseng. Why we twit-
Bernard, Eugene C Johnsen, and Gene A Shelley. Com- ter: understanding microblogging usage and communi-
paring two methods for estimating network size. Human ties. In Proceedings of the 9th WebKDD and 1st SNA-
Organization, 60(1):28, Nov 2001. KDD 2007 workshop on Web mining and social network
[10] H. Simon. Computers, Communication, and the Pub- analysis, page 56, 2007.
lic Interest, chapter Designing Organizations for an [20] Meeyoung Cha, Hamed Haddadi, Fabricio Benevenuto,
Information-Rich World, pages 38–52. Johns Hopkins and Krishna P. Gummadi. Measuring user influence in
Press, 1971. twitter: The million follower fallacy, 2010.
[11] T. Davenport and J. C Beck. The Attention Economy: [21] A. Barrat, M. Barthelemy, and A. Vespignani. Dynamical
Understanding the New Currency of Business. Harvard Processes On Complex Networks. Cambridge University
Business Press, 2002. Press, 2008.
[12] B. A. Huberman and F. Wu. The economics of attention: [22] Mark Newman. Networks: An Introduction. Oxford Uni-
maximizing user value in information-rich environments. versity Press, 2010.
In Proceedings of the 1st international workshop on Data
mining and audience intelligence for advertising, page 16,