Asemanticnetworkanalysisofcorporate
sustainabilitycommunicationinemerging
markets
Se Jung Park, Georgia State University
H...
Background
• Corporate Social Responsibility : embracing
sustainability into business strategy to gain
social benefits and...
Background
• Pohl (2006) noticed that corporate social
responsibility (CSR) represents the broad
spectrum of a company’s c...
The purpose of this study
• The literature on the CSR and global environmental
management has been biased in developed and...
Why Korea and China?
• Emerging consumer markets
• Increasing corporate power and economic
influence in the world
• Seriou...
Hofstede’s cultural dimension
Uncertainty Avoidance: “The extent to
which the members of a culture feel
threatened by ambi...
Research Questions
• RQ1: What are the semantic patterns of CER approaches
taken by Korean and Chinese large companies on ...
Mixed-Methods
• Data Collection
: Korean and Chinese corporations examined were sampled from the
country’s list of the top...
Findings
Centralityofsemanticnetwork
• Korean CER network (54.80%) was more centralized than
Chinese CER network (38.69%)....
Density of network
• Korean corporations had denser semantic network of CER
than Chinese corporations.
• Korea (Mean: 446....
KoreanCERSemanticNetwork
EgoNetworkof“Environmentin”KoreanCER
ChineseCERSemanticNetwork
EgoNetworkof“Environment”inChineseCER
KeythemesinKoreanCERNetwork
Risk minimization global
environmental issues
Commitment and responsibility for
environmental ...
KeythemesinChineseCERNetwork
Commitment to economy
and society
Resource conservation and
environmental responsibility
Regu...
Discussion & Conclusion
• The results imply that Korean corporations focus on presenting their
capability and pragmatic sk...
Discussion & Conclusion
• Korean firms were more strategical in articulating their
environmental initiatives, visions, per...
This study contributes to..
• The results provide theoretically meaningful insights for
evaluating corporate practices in ...
Limitation
• Different number of samples from each country was used
although this was due to discrepancies in their real
p...
Thank you!
1Dr. Nick Guldemond
The Micro Foundations of Triple Helix, Workshop
May 26-27 2014, Grenoble Ecole de Management
User Grou...
2Dr. Nick Guldemond
Road map
•Introduction: Grand societal health
challenges, user involvement and Living labs
•Research q...
3Dr. Nick Guldemond
Grand Societal Health Challenges
To maintain the health care
affordable, make it more
effective and or...
4Dr. Nick Guldemond
2007 EU
•average ~ 1:4
•differences between countries
2050 EU
•1:3 (NL)
•1:1.5 (Italy and Spain)
• EU ...
5Dr. Nick Guldemond
Netherlands
Prof.dr. Marina van Geenhuizen
6Dr. Nick Guldemond
medical curative model
community care
university hospital
local hospital
social (interconnected)
healt...
7Dr. Nick Guldemond
Stakeholder Complexity
Prof.dr. Marina van Geenhuizen
8Dr. Nick Guldemond
Users and Living labs
In the medical sector, there are more than
one user group:
•Patients, elderly pe...
9Dr. Nick Guldemond
User-involvement and Living Labs (Source: Almirall, Lee and Wareham, 2012)
10Dr. Nick Guldemond
Challenges of Living Labs:
Involvement of the right user groups
(motivation, capabilities)
Position...
11Dr. Nick Guldemond
Research questions and methodology
Research Questions
What are the characteristics of user-groups in ...
12Dr. Nick Guldemond
Character of Living Labs
Two operational levels (Følstad 2008)
• Open innovation networks or platform...
13Dr. Nick Guldemond
Preliminary set of LL critical factors (literature
Criterion Details
1.Involvement
of user groups
-Ad...
14Dr. Nick Guldemond
Case studies and user groups
1. Doornakkers (NL) real-life: Elderly of Turkish origin
2. Living Labs ...
15Dr. Nick Guldemond
Critical factors concerning users (1)
Doornakkers (Eindhoven-NL)
• eHealth/domotics, safety (maintain...
16Dr. Nick Guldemond
Critical factors concerning users (2)
i360 Royal College of Surgeons (Dublin-IRE)
• Healthcare/surgic...
17Dr. Nick Guldemond
Case studies: larger scale platforms
Pontes Medical (Amsterdam-Utrecht, NL)
• Health care and medical...
18Dr. Nick Guldemond
Answers to questions
Characteristics of user-groups in medical Living Labs:
• User groups are mainly ...
19Dr. Nick Guldemond
Critical factors in having user-groups
involved (Patient-oriented Living Labs)
1. Prior study of user...
20Dr. Nick Guldemond
Critical factors in having user-groups
involved (hospital/clinician oriented
Living Labs)
1. Trust cr...
21Dr. Nick Guldemond
Future lines of research
• To validate the outcomes using expert opinion.
• To increase the number of...
22Dr. Nick Guldemond
Thank you!
Prof.dr. Marina van Geenhuizen
Introduction to TU Delft – TPM
T
Prof. dr. Marina van Geenhuizen (TPM)
TU Delft
• TU Delft is in the city of Delft in The
Netherlands (European Union).
• It is the largest University of Technol...
TU Delft
Faculties
• Aerospace Engineering
• Applied Sciences
• Architecture and the Built Environment
• Civil Engineering...
Why a Challenge to come to TU Delft
for a Master Study?
TU Delft has risen to 42nd place in the global
reputation ranking ...
Mission
TU
Delft
Research
Valorization
Education
City of Delft
• Small city (95.000 inhabitants) but surrounded
by larger cities The Hague (Peace and Justice)
and Rotterda...
Faculty of Technology, Policy and
Management
Four MSc Programs
- Engineering and Policy Analysis
- Systems Engineering, Po...
Methods of Education
• Problem-oriented and geared towards design of problem
solutions (like in traffic and adoption of ne...
Requirements (TPM)
- BSc degree in a technical domain
- Cumulative Grade Point Average (CGPA) of at least 75% of
the scale...
July 3, 2014 10
Thank you !
Big Data and the Triple
Helix - a bibliometric
perspective
Martin Meyer*, Wolfgang Glanzel & Kevin Grant
*Kent Business Sc...
Purpose
• Big data has become the buzz word in recent years:
• topic of interest to a multitude of players
• be it governm...
Our Study
The Triple Helix Aspect
• Bibliometric study of TH indicators literature
• 110 papers, analysis of references cited
• 2 gr...
Triple Helix from a bibliometric
perspective
• Work at the heart
of the TH
• Cluster 1
• located at the heart
of the detai...
Triple Helix from a bibliometric
perspective
• The ‘neo-institutional’
side of the TH
• Science-technology
linkage
• Clust...
Triple Helix from a bibliometric
perspective
• The Neo-evolutionary Approach:
• Mutual information, entropy, and
sub dynam...
‘Big data’ – a bibliometric snapshot
• Based on 1,500 articles, letters, reviews and notes with BIG DATA as
topic or title...
Some Basic Stats
Research Areas Records %
COMPUTER SCIENCE 803 57.6
ENGINEERING 462 33.2
TELECOMMUNICATIONS 98 7.0
SCIENCE...
Some Basic Stats
Expected players visible
Countries/Territories Records %
USA 573 41.1
PEOPLES R CHINA 187 13.4
ENGLAND 91...
Searching for big
data
• Evolution of n of publications (left)
and citations (right) in WoS
Source: Thomson Reuter
Early s...
Removing ‘outliers’
Strong effect
shows the rapidly growing field
2011/12 onwards
Still strong influence of early papers
•...
Zooming in
•TOPIC: "BIG DATA“
•Timespan: All years.
•Refined by RESEARCH
AREAS:
• BUSINESS ECONOMICS
• INFORMATION SCIENCE...
Topics and Keywords
• Analysis based on DE and ID fields in WoS records
• Included all keywords/topics occurring more than...
Topics map
Service
Innovation
Data Privacy/
Politics
Retail,
Supply Chain &
Logistics
Social Media, Ethics & Philosophy
Mapping of Big Data Works
• Based on links of shared topics and references
• 187 papers
• 2881 references and terms
• 60 m...
Cluster analysis
Clusters
• Cluster 1: ‘possibilities and
challenges’: Big data and social
research (Psychology, TFSC, etc)
• Cluster 2:
In...
Cluster 1: ‘possibilities and challenges’: Big
data and social research (Psychology, TFSC,
etc)
AU TI- SO-
Bentley RA; O'B...
Cluster 2: Informetrics/Scientometrics
AU TI- SO-
Park HW; Leydesdorff L
Decomposing social and semantic networks in emerg...
Cluster 3: Big Data and the Media
AU TI- SO-
Bruns A; Highfield T; Burgess J
The Arab Spring and Social Media Audiences: E...
Cluster 4: Big Data as a Driver of
Change: ‘Challenges and IT Solutions’
AU TI- SO-
Rust RT; Huang MH
The Service Revoluti...
Cluster 5: Big Data and Geography
AU TI- SO-
DeLyser D; Sui D
Crossing the qualitative-quantitative divide II: Inventive
a...
Cluster 6: Big Data in the cloud:
Information systems related contributions
AU TI- SO-
Tien JM Big Data: Unleashing inform...
Cluster 7: Techniques to analyse Big
Data
AU TI- SO-
Janowicz K Observation-Driven Geo-Ontology Engineering TRANSACTIONS I...
Cluster 8: Big Data and Big Brother:
Cyber Surveillance
AU TI- SO-
Hu M Biometric ID Cybersurveillance INDIANA LAW JOURNAL...
Cluster 9: Big Data and Decision Support
Systems
AU TI- SO-
Demirkan H; Delen D
Leveraging the capabilities of service-ori...
Outlook
• New field, little work linking the various themes:
• BIG DATA the one key denominator
• Emerging differentiation...
SPEECH ACTS IN TELEVISED PRESIDENTIAL
DEBATES AND FACEBOOK MESSAGES:
THE CASE OF THE 2012 SOUTH KOREAN
PRESIDENTIAL ELECTI...
Purpose of the current study
 With the advent of social networking sites (SNSs),
ordinary individuals have opportunities ...
Speech acts
 Language use goes beyond the boundary of the
syntactic structure and its semantic meaning
 Language is used...
Televised presidential debates and
speech acts
 A few studies have attempted to understand how
debate participants use di...
Televised presidential debates and
speech acts
 The use of interrogatives can be perceived as an
aggressive tactic used b...
Televised presidential debates and
speech acts
 The presidential candidates during the 2004 U.S.
presidential debates fre...
Suggested hypotheses (part one)
 H1. Presidential candidates are more likely to use
constatives than any other type of sp...
Speech acts on candidates’ Facebook fanpages
 With respect to CMC messages, assertives are the dominant
type of speech ac...
Suggested hypotheses (part two)
 H5. Visitors to presidential candidates’ Facebook pages
are more likely to use assertive...
Method
 Samples
 the debate script was extracted for each candidate from
http://www.debates.go.kr: 609 sentences for Par...
Method
 Coding
Code Examples
Constatives “She doesn’t have any idea about economic democratization,” “He was t
oo gentle,...
Results
Speech acts Frequency Percentage
Constatives 933 67.4
Directives 35 2.5
Commissives 198 14.3
Expressives 53 3.8
In...
Results
Speech acts Frequency (%) Chi-square P
Park Moon
Constatives 623 583 .09 n.s.
Directives 113 164 9.92 <.01
Commiss...
Results
 Both candidates uttered more acclaims than any
other speech acts, consistent with the findings of
previous resea...
Results
 Moon’s fanpage visitors used more commissives and
directives than Park’s visitors.
 Moon’s visitors used more q...
Concluding remarks
 First, the candidates were most likely to employ clams
for truth (constatives), promises for the futu...
Introducing the
Oxford Internet Institute (OII)
Prof. Ralph Schroeder
• Social sciences department at
the University of Oxford
• Undertaking rigorous multi-
disciplinary research and
teaching ...
Taught Courses
• 50+ graduate students from wide variety of disciplinary backgrounds, and
from industry or government
• DP...
Michaelmas Hilary Trinity
Methods Social Research Methods and the
Internet Part I
Social Research Methods
and the Internet...
Two Options
• Digital Era Government and Politics
• Internet Economics
• Law and the Internet
• Online Social Networks
• L...
OII Research
• Topics covered across Governance and Democracy, Everyday
Life, Science & Learning, Network Economy, Shaping...
Other relevant projects
• Future Home Networks & Services (Ian Brown & Joss Wright):
researching and developing security f...
Research Examples
• People and Research
• Big Data: UK Government
• OxIS
• Political Science: Helen Margetts
• Geography: ...
Big Data:
UK Government Online
.
• JISC UK Web Domain
Dataset (30 Tb) of .uk ccTLD
from 1996-2010
• Here shows link struct...
 Data
 Internet Archives data of .uk back to 1996
 Annual crawls of .uk websites since 2013
 2.7 billion nodes, 40TB c...
Growth of subdomains
N.B. y-axis on log scale
Relative sector size on the web
Sectoral linking
2010
OII Faculty
Use by Age
(QH14 by QD1)
OxIS 2005: N=2,185; OxIS 2007: N=2,350; OxIS 2009: N=2,013
16
Which is more Important: Age or
Income?
Internet Users in Each Age-Income Category
(percents)
Age Groups
Income 14-44 45-6...
Use by Education (QH14 by QD14)
OxIS 2007: N=2,350; OxIS 2009: N=2,013 (Basic: N=901; Further: N=510; Higher: N=360).
Note...
Web 2.0 User Creativity & Production Online (QC10 and
QC31)
Current users. OxIS 2005: N=1,309; OxIS 2007: N=1,578; OxIS 20...
Helen Margetts ESRC Professorial Fellowship 2011-2014
The Internet, Political Science And Public Policy
Re-examining Colle...
•Social network
map of Bernie
Hogan’s FB ties,
Dec. 2008;
•Proof of concept
network that led
to creation of
NameGenWeb
Map...
Family
Local Friends
Three co-worker groups
Friends
Mark Graham: Total number of Wikipedia articles per 100,000 people
•Mark Graham & Bernie
Hogan’s project
investigates inequalities
in the creation of
knowledge.
• Map reveals uneven
spread ...
Sandra Gonzalez-Bailon
USENET Political Discussions (1999-2005)
0
2
4
6
8
x10000
09/1999 09/2000 09/2001 09/2002 09/2003 0...
Emotions and Public Opinion
Oxford e-Social Science Project
• Social shaping and
implications of e-Research
• Collaborative project with:
• SBS / InSI...
Source: Schroeder, R., Meyer, E.T. (2009). Gauging the Impact of e-Research in the Social Sciences. Paper presented
at the...
Source: Meyer, E.T., Park, H-W., Schroeder, R. (2009). Mapping Global e-Research: Scientometrics and Webometrics. Proceedi...
Source: Meyer, E.T., Schroeder, R. (2009). Untangling the Web of e-Research: Towards a Sociology of Online Knowledge. Jour...
Source: Meyer, E.T., Schroeder, R. (2009). Untangling the Web of e-Research: Towards a Sociology of Online Knowledge. Jour...
Source: Schroeder, R., Meyer, E.T. (2009). Gauging the Impact of e-Research in the Social Sciences. Paper presented at the...
For more information
see our website:
http://www.oii.ox.ac.uk
Twitter: @oiioxford
Big Data, Big Brother, and
Social Science
Ralph Schroeder
Collaborators:
Eric T. Meyer, Linnet Taylor, Josh Cowls, Greg Ta...
Overview
• Projects
• Questions
• Issues
• Definition
• How knowledge advances
• Examples
• Big Data Issues in Research an...
Accessing and Using Big Data to Advance
Social Science Knowledge
• Funded by Sloan Foundation
• Data sources
• 100+ interv...
See http://www.oii.ox.ac.uk/research/projects/?id=98
Data-driven economic models: challenges
and opportunities of big data
• Funded by Research Councils UK (RCUK),
New Economi...
Source: http://www.forbes.com/sites/davefeinleib/2012/06/19/the-big-data-landscape/
Source: Leonard John Matthews, CC-BY-SA (http://www.flickr.com/photos/mythoto/3033590171)
Spurious Correlations
Twitter-bots
OII master’s students Alexander Furnas and Devin Gaffney saw a large spike in then-US
presidential candidate ...
Google Images: Big Data
Source: Hill, K. (Feb 16, 2012). Forbes.com. Available at: http://www.forbes.com/sites/kashmirhill/2012/02/16/how-target-f...
113 240 278 367
558
1,195
1,538
2,350
3,960
6,787
7,276
9,010
-
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000
10,0...
Big data in the commercial world
• Commercial uses are: ‘in house’,
‘outsourced own data’, ‘data analysis as a
consultancy...
Definition
• ‘Big data’
– the advance of knowledge via a leap in the scale
and scope in relation to a given object or
phen...
Computational Manipulability?
• ‘the distinctiveness of the network of mathematical
practitioners is that they focus their...
Research computing
The Grid
Supercomputing
Clouds
Big Data
Web 2.0
Digital Objects and their Referents
Digital Object
(Examples: Twitter,
Tesco Loyalty card
information
Real World
(People /...
Representing
Manipulating
Limits
Digital Data
010101 Knowledge
Uses and Limits
• Big data research uses (academic, commercial, government) are limited to
the exploitation of suitable ob...
113 240 278 367
558
1,195
1,538
2,350
3,960
6,787
7,276
9,010
-
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000
10,0...
Platform Paper Size of Data in relation to
phenomenon investigated
Theoretical
question/practical aim
Key findings
Faceboo...
Example 1:
Search engine behaviour
Waller’s analysis ofAustralian Google Users
Key findings:
- Mainly leisure
- > 2% conte...
?
?
?
?
?
?
?
?
?
“Surprisingly, the distribution of
types of search query did not vary
significantly across the different...
Example 2:
Large-scale text analysis
Michel et al. ‘culturomic’ analysis of 5 Million Digitized Google
Books and Heuser & ...
J Michel et al. Science 2011;331:176-182
Example 3:
Social network or news?
Kwak et al.’s analysis ofTwitter
Key findings:
- 1.47 billion social relations
- 2/3 of...
(Big) data definition enables
pinpointing impacts and threats
• ‘Google Plus may not be much of a competitor to Facebook a...
‘Big data‘ for understanding society
• Real-time transactional data (unlike survey
data, traditional staple of social scie...
Social Science and Big Data
Research
• Dominated by social media
• Issues of ‘whole universe’
– What population, offline a...
Scientificity and Big Data: Pro and
Con
• Pro
– Replicability, extension to new domain
– ‘Total’ datasets, ‘whole universe...
Ethical and Social Issues in Big Data Research
• Objects with ‘total’ knowledge (universes)
– Danger is inferring behaviou...
Other positions on Big Data
Implications 1
• Mayer-Schoenberger and Cukier, boyd and Crawford argue that not
all informati...
Other Positions on Big Data Implications 2
• Savage and Burrows: ask are commercial data outpacing
social science?
• Boyd ...
Consumer (and gov’t) Big Data
• Consumer data and privacy (ie. Target pregnancy case)
– Solution: data protection
• Consum...
Big Data and Policy
• Probabilistic rather than ‘causal’ commercial and
government uses of data (ie. profiling) - only pro...
Future of Big Data Research
• Difference commercial versus academic world is that
knowledge provides competitive advantage...
Outlook and Implications
• There is an overlap between real world research and
the world of academic research which is clo...
Additional readings and references
Bond, Robert et al. (2012). ‘A 61-million-person experiment in social influence and pol...
Project Papers
Schroeder, Ralph (Forthcoming). ‘Big Data: Towards a More Scientific Social Science and Humanities’ in Mark...
Oxford Internet Institute
With support from:
Ralph Schroeder
ralph.schroeder@oii.ox.ac.uk
http://www.oii.ox.ac.uk/people/?...
Understanding “Wedge-Driving” Rumors
Online during a Political Crisis: Insights
from Twitter Analyses during Korean
Saber ...
Rumors Revisited
• Unofficial Information Sharing in Social
Media
• Unofficial Information = Rumors =
Representation of bo...
Goals of the Study
• Theoretically: Understanding social media
rumormongering as a contentious process of
collectively con...
Public Opinions
• Public Opinions: (1) citizen responses as
opposed to governing actors; (2) expressed
openly instead of p...
Opinion Polling…
• A top-down, institutionalized construction
of public opinions
• Quantitative, limited conveyance of opi...
Rumors: Improvised Public Opinions
• Alternative indicators of opinion climate
• Bottom-up, unstructured construction of
s...
Textual Analysis of Rumors
• Social Psychology of Rumors
• Textual Analysis of Rumors
- Only a few studied due to the lack...
Wedge-Driving (WD) Rumors
• 3 rumor types during a crisis: wish, dread, WD
• WD rumors: a moniker for unverified propositi...
Empirical Research Questions:
To what extent does rumoring happen
in social media when a society faces a
social/political ...
Case: Korean Saber Rattling
2013
• Rumormongering = uncertainty (ambiguous
situation) x anxiety (issue importance)
• Saber...
Small-Scale Content Analysis
• Quota sampling of 2,500 non-redundant,
unique tweet messages (2,352 after
filtering) from a...
Content Analysis
• Dummy coding: (1) informational ambiguity
(84.5% agreement), (2) propositional statement
(88.9% agreeme...
Semantic Network Analysis
• Words selected based on Bonferroni-
adjusted z-tests of word frequency
comparisons among the 3...
General Results
• 25% NR message (62 words), 36.4% WD
messages (99 words), and 38.6% GR
messages (41 words)
• Two centrali...
Non-Rumor Semantic Network
Non-Rumor Semantic Network
Non-Rumor Semantic Network
Non-Rumor Semantic Network
Non-Rumor Semantic Network
NR network highlights…
• Formal, top-down responses to the threat,
in a broader geopolitical context.
 SK’s political and...
WD Semantic Network
WD network highlights
• Derogatory themes:
 Defaming historic or current politicians
(C1), even a public figure in a non-...
GR semantic network
GR network
GR network
GR semantic network
GR network highlights…
• Bottom-up reaction to the threat
 the public’s curiosity about the NK’s
readiness of kinetic war...
Discussion & Conclusions
• Nontrivial portion of spontaneous, less-
than-rational public responses to social or
political ...
• Non-rumors: similar to institutional polling
(e.g. Gallup questionnaire)
• General-rumors: derivative of the news
agenda...
Limitation & Future Research
• Threw away a large amount of available
data due to limited methods
• Needs to incorporate a...
A social network framework to analyze the
cultural contents of Kpop across countries
Ji-Young Park & Ji-Young Kim
(PhD stu...
Contents
• Cultural phenomenon of the Korean wave
• Variety of Data procedure
- Data preparation
- Data process
• Social n...
Cultural phenomenon of the Korean wave
• Hallyu(한류: Korean Wave) is a neologism referring to the
increase in the popularit...
Cultural phenomenon of the Korean wave
• Cultural exports such as Hallyu (“Korean Wave”) embody the
global influence of lo...
Web 1.0 Korean Wave Web 2.0 Korean Wave
Period Early 2000s 2010s
Genre Mostly TV dramas Multiple Contents
(e.g. K-pop, Onl...
• This study focuses on Kpop and a Korean
rapper Psy’s Gangnam Style (GS)
Research Questions
• What is the communication patterns among
international fans of Kpop across countries ?
• Various kinds of online data are used in current paper.
• The big data-based analysis programs, including the
Webometric...
• (1) Web documents on Korean singers
• (2) Visibility of Korean singers at popular social
media sites
• (3) Communication...
Social Network Analysis Framework Data procedure Method SNA tool
(1) Web documents on Korean singers
- Scrape keyword(Kore...
Social network analysis framework
• (1) Web documents on Korean singers
- Webonaver, Webogoogle
Social network analysis framework
• (1) Web documents on Korean singers
- Webonaver as a scrapper tool
-NaverScrapper - Sc...
Social network analysis framework
• (1) Web documents on Korean singers
- WeboGoogle as a scrapper tool
-WeboGoogle - Scra...
Social network analysis framework
• (1) Web documents on Korean singers
- WeboGoogle as a scrapper tool
- The results base...
Social network analysis framework
• (2) Visibility of Korean singers at popular
social media sites
-Twitter, Facebook
Usin...
Social network analysis framework
• (3) Communication patterns among
international fans of Kpop across countries
Webometri...
Social network analysis framework
• (3) Communication patterns among
international fans of Kpop across countries
• Using w...
Social network analysis framework
• A user-to-user network was constructed to reveal hidden
relationships between commente...
Social network analysis framework
• In terms of the geographical distribution of
commenters, the U.S. had the largest numb...
Results
• This structural difference between the NC and the NSCN can
be explained in part by the nature of YouTube.
• In t...
Figure1. Commentary network in August
Gangnam Style Communication Networks on Youtube
chain shape reflecting a circle
•.
Figure 2. Subscriptions to a common network in August
Gangnam Style Communication Networks on Youtube
hub-and-spoke top...
• The structural pattern of the NC
• Correlation analysis of common networks
• These results indicate that frequent replie...
In terms of the structural pattern of the NSCN,
• According to the independent sample t-test, U.S. (N = 158) and non-
U.S....
Discussion & Implication
• Asian popular music has grown rapidly, particularly in the
U.S. and European countries, but suc...
An analysis of Twitter communication
on Organic products in Mexico and Korea
using webometrics method.
G.CD. Xanat V. Meza...
Objectives
• The present study compares social media resources for organic
products between Mexico and Korea in the Twitte...
Literature Review Cross cultural research and SNS.
• This study will apply a framework by Marcus & Gould (2001),
which is ...
Method Webometrics.
“The study of web-based content
with primarily quantitative methods
for social science research goals ...
Method Semantic analysis.
• Analyses semantic relationships between concepts (Sowa, 1987).
• In the present study, the uni...
Method Data collection procedures.
• Hashtags for “Organic”:
• Organico (in spanish)
• 유기농 (in korean)
• The process:
• Co...
Results
RQ1.What is the diffusion path of social media resources for
organic products in Mexico and Korea through Twitter?...
Results RQ1.1.How are the networks changing through time?
Results RQ1.1.How are the networks changing through time?
Results RQ1.1.How are the networks changing through time?
0
2000
4000
6000
8000
10000
January
February
March
April
May
Jun...
Results RQ1.1.How are the networks changing through time?
Correlations for Mexico
Vertices Edges
Maximum Geodesic
Distance...
Results RQ1.2. Who are influential players in
diffusing organic products on Twitter?
Results RQ1.2. Who are influential players in
diffusing organic products on Twitter?
Results RQ1.2. Who are influential players in
diffusing organic products on Twitter?
Indegree Centrality value Type of use...
Results RQ1.2. Who are influential players in
diffusing organic products on Twitter?
ALTERNATIVE MEDIA
1
POLITICIAN
2
BUSI...
Results RQ1.2. Who are influential players in
diffusing organic products on Twitter?
Indegree Centrality value Type of use...
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
인문학육성기금 Cyber emotions research center june_2014
Upcoming SlideShare
Loading in …5
×

인문학육성기금 Cyber emotions research center june_2014

840 views
731 views

Published on

영남대 사이버감성연구소

Published in: Technology, Business
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
840
On SlideShare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
0
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

인문학육성기금 Cyber emotions research center june_2014

  1. 1. Asemanticnetworkanalysisofcorporate sustainabilitycommunicationinemerging markets Se Jung Park, Georgia State University Hongmei Li, Georgia State University Han Woo Park, YeungNam University
  2. 2. Background • Corporate Social Responsibility : embracing sustainability into business strategy to gain social benefits and create business value (Dauvergne & Lister, 2012). • Corporate Sustainability: A firm's awareness of environmental protection issues and its incorporation of ecological concern and sustainable development for long-term growth (Lu & Li, 2009).
  3. 3. Background • Pohl (2006) noticed that corporate social responsibility (CSR) represents the broad spectrum of a company’s corporate culture, including values, beliefs, attitudes, and norms. • Culture has been regarded as critical element in business ethical decision-making and PR strategies (Kim & Kim, 2010) • However, there is little empirical data on the relation between cultural and PR approaches
  4. 4. The purpose of this study • The literature on the CSR and global environmental management has been biased in developed and Western countries such as those in North America and Europe (Leonidou & Leonidou, 2011). • Little studies have addressed corporate CER approaches in new media setting though new media is the main platform to reach widest consumers. • This study investigates how large firms in Korea and China employ CER communication through their websites to better understand the major approaches taken by these firms to disclose their CER principles and compares the two countries in terms of their cultural similarities and differences through mixed method approaches.
  5. 5. Why Korea and China? • Emerging consumer markets • Increasing corporate power and economic influence in the world • Serious environmental problems and lower concerns on environmental issues • Relatively low efforts and perception for environmental sustainability
  6. 6. Hofstede’s cultural dimension Uncertainty Avoidance: “The extent to which the members of a culture feel threatened by ambiguous or unknown situations and have created beliefs and institutions that try to avoid.” Individualism: “the degree of interdependence a society maintains among its members.” South Korea is higher uncertainty avoiding culture than China. Both countries are collectivistic culture.
  7. 7. Research Questions • RQ1: What are the semantic patterns of CER approaches taken by Korean and Chinese large companies on their websites? • RQ2:What are the key themes in CER communication of Korean and Chinese large companies on their websites? • RQ3: How do cultural differences shape these CER communication strategies?
  8. 8. Mixed-Methods • Data Collection : Korean and Chinese corporations examined were sampled from the country’s list of the top 50 largest corporations in terms of revenue. A total of 44 Korean firms provided CER-related information, whereas 32 Chinese firms provided it • Semantic Network Analysis • This study employed semantic network analyses based on the top 100 frequently used key words. A co-word analysis and cluster analysis (CONCOR) were conducted for specifying key themes from texts. • “It examines the relationships among a system's components based on the shared meanings of symbols (Doerfel & Barnett, 1999).” FullText, a network analysis tool was used (http://www.leydesdorff.net/software/fulltext/). • Qualitative content analysis • Identified key CER principles and the relation of cultural values and prominent themes.
  9. 9. Findings Centralityofsemanticnetwork • Korean CER network (54.80%) was more centralized than Chinese CER network (38.69%). • Prominent words in Korean CER: Management (58.715), we(49.599), green (44.522), our (37.743), energy(35.827), system (33.55), environment (33.045) • Prominent words in Chinese CER: Energy (45.691), company (45.691), we (34.87), development (34.62), china (32.03), our (26.859), management (26.859)
  10. 10. Density of network • Korean corporations had denser semantic network of CER than Chinese corporations. • Korea (Mean: 446.510, SD: 865.747) • China (Mean: 163.929, SD: 212.001)
  11. 11. KoreanCERSemanticNetwork
  12. 12. EgoNetworkof“Environmentin”KoreanCER
  13. 13. ChineseCERSemanticNetwork
  14. 14. EgoNetworkof“Environment”inChineseCER
  15. 15. KeythemesinKoreanCERNetwork Risk minimization global environmental issues Commitment and responsibility for environmental change Improvement of eco-technology Efficient use of resource and suitability management system Endorsement for green facility Employee education & workplace security Internal and international management system of hazardous substance Collective efforts to embracing environmentalism
  16. 16. KeythemesinChineseCERNetwork Commitment to economy and society Resource conservation and environmental responsibility Regulation on management of environmental protection & consumer right Development of green products & supports to government Improvement of local economies and awareness on global environment Implement of national policies & social value Environmental concerns Advanced facilities
  17. 17. Discussion & Conclusion • The results imply that Korean corporations focus on presenting their capability and pragmatic skills to resolve environmental problems as economic powers, while Chinese corporations are more concerned with their brand image as social responsible and engagement with stakeholders. • Korean and Chinese corporations framed CER principles and practices differently: While the Korean corporations focused on promoting eco-friendly technologies as a competitive strategy and frequently used performance-related terms, the Chinese corporations employed more collectivistic appeal such as commitment to local and global communities and partnerships with NGOs and stakeholders. • Hofstede’s cultural dimensions on collectivism explain the similar approach of CER in both countries.
  18. 18. Discussion & Conclusion • Korean firms were more strategical in articulating their environmental initiatives, visions, performance, activities, and environmental concerns with detailed reports in comparison with Chinese companies. • This can be explained by Korea’s high uncertainty avoiding culture that corresponds to the institutions’ concerns on surrounding environments and future condition.
  19. 19. This study contributes to.. • The results provide theoretically meaningful insights for evaluating corporate practices in Asia in terms of communicating environmental management strategies and their principles in the context of new media. Given the lack of environmental management research in the business communication domain, this study contributes to the literature by empirically analyzing East Asian firms' campaign performance. • In addition, the study provides important methodological implications for the analysis of corporate websites and demonstrates mixed methods to extract and analyze a large size of texts in a systemic way.
  20. 20. Limitation • Different number of samples from each country was used although this was due to discrepancies in their real performance. Another explanation for this may be related to the data including only English versions of websites. • This study considers only websites for examining CER, but many firms now make increasingly active use of social media such as Twitter and Facebook for marketing purposes.
  21. 21. Thank you!
  22. 22. 1Dr. Nick Guldemond The Micro Foundations of Triple Helix, Workshop May 26-27 2014, Grenoble Ecole de Management User Groups in Triple Helix Interaction: The Case of Living Labs in Health Innovation Marina van Geenhuizen* and Nick Guldemond** * TU Delft **University Medical Centre Utrecht
  23. 23. 2Dr. Nick Guldemond Road map •Introduction: Grand societal health challenges, user involvement and Living labs •Research question •Methodology: literature study and six case studies •Preliminary list of critical factors •Results of case studies •Conclusions on critical factors and future research steps
  24. 24. 3Dr. Nick Guldemond Grand Societal Health Challenges To maintain the health care affordable, make it more effective and oriented towards persons in a situation of ageing population and shrinking budgets!!
  25. 25. 4Dr. Nick Guldemond 2007 EU •average ~ 1:4 •differences between countries 2050 EU •1:3 (NL) •1:1.5 (Italy and Spain) • EU average ~ 1:2 2050 China •average ~ 1:<1 Population aged (>65) in proportion to working population (18-65) Prof.dr. Marina van Geenhuizen
  26. 26. 5Dr. Nick Guldemond Netherlands Prof.dr. Marina van Geenhuizen
  27. 27. 6Dr. Nick Guldemond medical curative model community care university hospital local hospital social (interconnected) health perspective community care advanced local care centres high specialized cure
  28. 28. 7Dr. Nick Guldemond Stakeholder Complexity Prof.dr. Marina van Geenhuizen
  29. 29. 8Dr. Nick Guldemond Users and Living labs In the medical sector, there are more than one user group: •Patients, elderly people, etc. •Family doctors •Medical staff in clinics •Clinics Why involvement of user groups/customers in design (co-creation)? The design process turns out to be quicker and more effective, like in design of artificial limbs (patients) and of surgery room equipment. Living Labs are one way to involve user groups/customers in innovation
  30. 30. 9Dr. Nick Guldemond User-involvement and Living Labs (Source: Almirall, Lee and Wareham, 2012)
  31. 31. 10Dr. Nick Guldemond Challenges of Living Labs: Involvement of the right user groups (motivation, capabilities) Positioning them in the network, given the dynamic stakeholder situation of which the Triple Helix partners (academia, industry and government) are only a few (also, insurance companies, registration authorities, venture capitalists, ngos etc.)
  32. 32. 11Dr. Nick Guldemond Research questions and methodology Research Questions What are the characteristics of user-groups in Living Labs? In which ways are Triple Helix partners active in Living Labs and can user-groups in interaction with them contribute to bringing new technology to market? Methodology: • Evaluation of the literature: critical factors in founding and managing Livings Labs • 6 in-depth case studies of medical Living Labs (multiple data sources) Prof.dr. Marina van Geenhuizen
  33. 33. 12Dr. Nick Guldemond Character of Living Labs Two operational levels (Følstad 2008) • Open innovation networks or platforms in a city/region • Real-life physical setting used for co-creation and testing with strong involvement of user groups Despite differences in size, setting, organization, driving actors, etc. three common characteristics: • An early involvement of user groups • A physical and/or social environment representing real-life • Open networks of stakeholders sharing the desire to support a better/quicker take up of inventions Prof.dr. Marina van Geenhuizen
  34. 34. 13Dr. Nick Guldemond Preliminary set of LL critical factors (literature Criterion Details 1.Involvement of user groups -Adequate model of involvement -Selection of users (motivation and capabilities needed) 2.Composition & management of the network -Involvement of all relevant actors to create vertical cooperation in the value chain and horizontal cooperation (scale economies). -Avoiding a too many partners, avoiding dominance of a powerful one and strong interdependency between powerful partners -Increasing openness and neutrality, including trust, to avoid one powerful partner to play a ‘key role’ deterring other partners to participate 3.Structured process -Working with a transparent ‘funnel’ or other innovation model -Working with clear go/no-go decisions 4.Role of ICT -Sufficient use of ICT in monitoring and analysis of user response in the design processes -ICT should not be the main driver, unless its adoption is subject of analysis, like in ambient assisted living 5.Operational management -Quality management of the networks is required, enabling the balancing of partners’ interests and managing expectations (and trust) - Transparency of distribution of tasks and cost/benefits over the partners 6.Practical requirements -Ethics/law: sufficient attention for ethical/legal issues, like users’ privacy and legal liability in case of failure -Intellectual property (IP): Sufficient attention necessary in early stage.
  35. 35. 14Dr. Nick Guldemond Case studies and user groups 1. Doornakkers (NL) real-life: Elderly of Turkish origin 2. Living Labs Amsterdam (NL) real-life: Elderly , housing foundation 3. i360 Royal College of Surgeons (Ireland) real-life: Medical staff (surgeons) 4. Medical Field Lab (NL) platform+real-life: Mix of users 5.Pontes Medical (NL) platform+real-life: Mix of users 6. Healthcare Innovation Lab (DK) real-life: Hospitals, clinicians, patients
  36. 36. 15Dr. Nick Guldemond Critical factors concerning users (1) Doornakkers (Eindhoven-NL) • eHealth/domotics, safety (maintain independent living) • Users: elderly from Turkish origin (isolated community) • Role of users: rather passive (sometimes active) • Triple Helix: disconnected from university • Success factor users: preparation study of needs; trust creation (coach of Turkish origin); ICT well managed Living lab Amsterdam-NL • eHealth/domotics, safety (maintain independent living) • Users: mixed elderly (also social housing foundation) • Role of users: manifold (designers, subjects, storytellers) • Triple Helix: business weakly involved; universities strongly involved (co-design, broader research on needs) • Success factors users: trust creation prior to project start; mixed methods in learning, multidisciplinary; ICT well managed; more attention needed for user values Prof.dr. Marina van Geenhuizen
  37. 37. 16Dr. Nick Guldemond Critical factors concerning users (2) i360 Royal College of Surgeons (Dublin-IRE) • Healthcare/surgical technology • Users: medical staff hospitals (surgeons) • Role of users: user-driven model • Triple Helix: active role for university, but government weakly involved; active reduction of TH gaps. • Success factors users: trust between partners, flexibility of users in shift from network to company
  38. 38. 17Dr. Nick Guldemond Case studies: larger scale platforms Pontes Medical (Amsterdam-Utrecht, NL) • Health care and medical technology (selected) • Users: Medical staff, care professionals, hospitals, patients, firms • Role of users: user-driven model (clinic driven) • Triple Helix: strongly connected and active reduction of TH gaps • Success factors users: protection of IO (clinicians, companies) Healthcare Innovation Lab (Copenhagen, DK) • New services, and organization and care concepts (e-health) and a methodology of user driven innovation (using simulation lab) • Users: Hospitals, clinicians, patients • Role of users: highly interactive in simulation lab • Triple Helix: strongly connected, but business weakly connected, and active reduction of TH gaps • Success factors users: selection of users on capabilities (simulation), trust between partners, passionate leadership Prof.dr. Marina van Geenhuizen
  39. 39. 18Dr. Nick Guldemond Answers to questions Characteristics of user-groups in medical Living Labs: • User groups are mainly patient-oriented (care/ treatment) or hospital/clinicians-oriented (facilities) • Their involvement may include various methods: co- design, story-telling, scenario-thinking, co-simulation Ways in which Triple Helix partners are active in Living Labs: • In Living Labs on e-health for elderly, either the university or industry tend to be weakly involved • In Living Labs for broader medical care/cure and hospital facilities, all three TH actors tend to be actively involved.
  40. 40. 19Dr. Nick Guldemond Critical factors in having user-groups involved (Patient-oriented Living Labs) 1. Prior study of user needs 2. Trust creation (eventually prior to project) and role models and coaches based on familiarity 3. Manifold inputs and multidisciplinary approach: co-design, story-telling, scenario-thinking, etc. 4. Attention for user values: ICT dependency, privacy, individuality 5. Moderate ‘dosage’ of new ICT 6. Passionate leadership for inspiration
  41. 41. 20Dr. Nick Guldemond Critical factors in having user-groups involved (hospital/clinician oriented Living Labs) 1. Trust creation between users and other partners 2. Flexibility in shift to new concepts, i.e. from network to company 3. Protection of IO of users (clinicians, companies) 4. Selection on capabilities of users 5. Passionate leadership
  42. 42. 21Dr. Nick Guldemond Future lines of research • To validate the outcomes using expert opinion. • To increase the number of Living Labs and to analyze them quantitatively (fuzzy set analysis): pattern recognition, causal structures, etc. • To compare medical Living Labs with Living Labs in other domains. • To determine what success of Living Labs would mean and how it can be measured (so far merely by process variables).
  43. 43. 22Dr. Nick Guldemond Thank you! Prof.dr. Marina van Geenhuizen
  44. 44. Introduction to TU Delft – TPM T Prof. dr. Marina van Geenhuizen (TPM)
  45. 45. TU Delft • TU Delft is in the city of Delft in The Netherlands (European Union). • It is the largest University of Technology in the Netherlands (10,500 bachelor students and 6,650 master students in 2012) Founded in 1842 as a Royal Academy for Engineering
  46. 46. TU Delft Faculties • Aerospace Engineering • Applied Sciences • Architecture and the Built Environment • Civil Engineering and Geosciences • Electrical Engineering, Mathematics and Computer Sciences • Industrial Design • 3-ME (Mechanical, Maritime and Material Engineering) • Technology, Policy and Management
  47. 47. Why a Challenge to come to TU Delft for a Master Study? TU Delft has risen to 42nd place in the global reputation ranking list of the World Reputation Rankings of Times Higher Education magazine. TU Delft is now the highest ranked Dutch university and the third highest European university of technology. Three other Dutch universities are ranked in this top 100.
  48. 48. Mission TU Delft Research Valorization Education
  49. 49. City of Delft • Small city (95.000 inhabitants) but surrounded by larger cities The Hague (Peace and Justice) and Rotterdam (World Port City) • Historical city center (houses from late middle ages and older) • Amsterdam at one hour drive, Brussels at two hours car drive • Paris and London also pretty nearby (45 minutes flight).
  50. 50. Faculty of Technology, Policy and Management Four MSc Programs - Engineering and Policy Analysis - Systems Engineering, Policy Analysis and Management - Transport, Infrastructure and Logistics - Management of Technology
  51. 51. Methods of Education • Problem-oriented and geared towards design of problem solutions (like in traffic and adoption of new technology) • Much group work (except for MSc thesis) • MSc often in internship (company, policy institute) • Strongly multidisciplinary (e.g. sustainable energy, health technology and medical care, water works)
  52. 52. Requirements (TPM) - BSc degree in a technical domain - Cumulative Grade Point Average (CGPA) of at least 75% of the scale maximum - Proof of English language proficiency So Welcome at TU Delft, TPM!! For further information: www.admissions.tudelft.nl E: Internationaloffice-tbm@tudelft.nl
  53. 53. July 3, 2014 10 Thank you !
  54. 54. Big Data and the Triple Helix - a bibliometric perspective Martin Meyer*, Wolfgang Glanzel & Kevin Grant *Kent Business School, University of Kent, Canterbury CT2 7PE, United Kingdom, m.s.meyer@kent.ac.uk
  55. 55. Purpose • Big data has become the buzz word in recent years: • topic of interest to a multitude of players • be it government or industry, academics or the public at large • This presentation will offer a bibliometric perspective • We analyze the emergent literature in the field • Our analysis will offer a general overview of developments and then zoom in​ focusing​ ​on areas of particular interest • publication activity in certain domains are focused on particular themes. • outlook as to what strongly emergent topics are
  56. 56. Our Study
  57. 57. The Triple Helix Aspect • Bibliometric study of TH indicators literature • 110 papers, analysis of references cited • 2 groups emerged: • Neo-evolutionary (mostly Leydesdorff and colleagues) • Neo-institutional (Etzkowitz, Leydesdorff)
  58. 58. Triple Helix from a bibliometric perspective • Work at the heart of the TH • Cluster 1 • located at the heart of the detailed network map of papers. • reaches out to both groups almost equally.
  59. 59. Triple Helix from a bibliometric perspective • The ‘neo-institutional’ side of the TH • Science-technology linkage • Cluster 8: mostly related to patent citation indicators to measure S&T linkage or discuss their usefulness. • Cluster 3: very closely aligned to the work on patent citation analysis as described above. • Entrepreneurial universities and university patenting • Cluster 4 is focused on the entrepreneurial university and ways of capturing researchers’ entrepreneurial and collaborative activity.
  60. 60. Triple Helix from a bibliometric perspective • The Neo-evolutionary Approach: • Mutual information, entropy, and sub dynamics • Cluster 2: approaches to capture triple helix relations in terms of information and communication flows and identify their knowledge bases. • Evolutionary Thinking & Knowledge Spill-overs • Cluster 7 : evolutionary theorising as well as the geography of innovation, especially on regional innovation systems and knowledge spill-overs. • Cluster 5 (closely related to Cluster 2 as well as 6) extends this perspective towards a framework for empirical research. • Cluster 6: innovation as an interactive process, leading from user-producer interactions to a national system of innovation, work on the intellectual and social organisation of the sciences as increasingly an organised and controlled knowledge production system
  61. 61. ‘Big data’ – a bibliometric snapshot • Based on 1,500 articles, letters, reviews and notes with BIG DATA as topic or title • Based on WoK SSCI/SCI indices • No claim that study is exhaustive but opens up a view on what kind of scholarly literature is currently associated with the Big Data label • We will be zooming in even further and look at a subset of Social Science / Information Science related works that could be potentially linked to TH indicators works
  62. 62. Some Basic Stats Research Areas Records % COMPUTER SCIENCE 803 57.6 ENGINEERING 462 33.2 TELECOMMUNICATIONS 98 7.0 SCIENCE TECHNOLOGY OTHER TOPICS 66 4.7 BUSINESS ECONOMICS 64 4.6 INFORMATION SCIENCE LIBRARY SCIENCE 56 4.0 OPTICS 45 3.2 PHYSICS 33 2.4 BIOCHEMISTRY MOLECULAR BIOLOGY 31 2.2 MATHEMATICS 29 2.1 • Big data covered in obvious research areas
  63. 63. Some Basic Stats Expected players visible Countries/Territories Records % USA 573 41.1 PEOPLES R CHINA 187 13.4 ENGLAND 91 6.5 GERMANY 66 4.7 AUSTRALIA 48 3.4 CANADA 47 3.4 JAPAN 45 3.2 SOUTH KOREA 38 2.7 NETHERLANDS 35 2.5 ITALY 33 2.4 FRANCE 29 2.1 SPAIN 29 2.1 INDIA 28 2.0 SWITZERLAND 25 1.8 TAIWAN 23 1.7 POLAND 16 1.1 SINGAPORE 15 1.1
  64. 64. Searching for big data • Evolution of n of publications (left) and citations (right) in WoS Source: Thomson Reuter Early study making reference to ‘big data sets’: “DnaSP, DNA polymorphism analyses by the coalescent and other methods” by Rozas, J; Sanchez-DelBarrio, JC; Messeguer, X; Rozas, R in BIOINFORMATICS (2003, 10.1093/bioinformatics/btg359) Strong effect: 3707 cites Next highest cited article; 120 citations
  65. 65. Removing ‘outliers’ Strong effect shows the rapidly growing field 2011/12 onwards Still strong influence of early papers • Evolution of n of publications (left) and citations (right) in WoS
  66. 66. Zooming in •TOPIC: "BIG DATA“ •Timespan: All years. •Refined by RESEARCH AREAS: • BUSINESS ECONOMICS • INFORMATION SCIENCE • LIBRARY SCIENCE • OPERATIONS RESEARCH MANAGEMENT SCIENCE • PSYCHOLOGY • COMMUNICATION • BEHAVIORAL SCIENCES • GEOGRAPHY • GOVERNMENT • LAW • SOCIAL SCIENCES OTHER TOPICS • SOCIOLOGY • SOCIAL ISSUES • 187 papers from Web of Science Core Collection
  67. 67. Topics and Keywords • Analysis based on DE and ID fields in WoS records • Included all keywords/topics occurring more than once • Total: 54 across the 136 papers that contained relevant fields • Triple Helix occurred 6 times • Normalised dataset (Jaccard) • Mapped in Pajek (Kamada Kawai) • Big data by far the most frequent term and by default at the centre of field: • Signs of emergent differentiation
  68. 68. Topics map Service Innovation Data Privacy/ Politics Retail, Supply Chain & Logistics Social Media, Ethics & Philosophy
  69. 69. Mapping of Big Data Works • Based on links of shared topics and references • 187 papers • 2881 references and terms • 60 most linked papers mapped in Pajek • Person’s cluster algorithm applied: • 9 clusters
  70. 70. Cluster analysis
  71. 71. Clusters • Cluster 1: ‘possibilities and challenges’: Big data and social research (Psychology, TFSC, etc) • Cluster 2: Informetrics/Scientometrics • Cluster 3: Big Data and the Media • Cluster 4: Big Data as a Driver of Change: ‘Challenges and Solutions’ (Mkt, Transp, IT related services) • Cluster 5: Big Data and Geography • Cluster 6: Big Data in the cloud: Information systems related contributions • Cluster 7: Techniques to analyse Big Data • Cluster 8: Big Data and Big Brother: Cyber Surveillance • Cluster 9: Big Data and Decision Support Systems
  72. 72. Cluster 1: ‘possibilities and challenges’: Big data and social research (Psychology, TFSC, etc) AU TI- SO- Bentley RA; O'Brien MJ; Brock WA Mapping collective behavior in the big-data era BEHAVIORAL AND BRAIN SCIENCES Boyd D; Crawford K CRITICAL QUESTIONS FOR BIG DATA Provocations for a cultural, technological, and scholarly phenomenon INFORMATION COMMUNICATION & SOCIETY Tangherlini TR; Leonard P Trawling in the Sea of the Great Unread: Sub-corpus topic modeling and Humanities research POETICS Huang TL; Van Mieghem JA Clickstream Data and Inventory Management: Model and Empirical Analysis PRODUCTION AND OPERATIONS MANAGEMENT Ballings M; Van den Poel D Customer event history for churn prediction: How long is long enough? EXPERT SYSTEMS WITH APPLICATIONS Enjolras B Big Data and social research: New possibilities and ethical challenges TIDSSKRIFT FOR SAMFUNNSFORSKNING Miller AR; Tucker C Health information exchange, system size and information silos JOURNAL OF HEALTH ECONOMICS Kern ML; Eichstaedt JC; Schwartz HA; Park G; Ungar LH; Stillwell DJ; Kosinski M; Dziurzynski L; Seligman MEP From "Sooo Excited!!!" to "So Proud": Using Language to Study Development DEVELOPMENTAL PSYCHOLOGY Jun SP; Yeom J; Son JK A study of the method using search traffic to analyze new technology adoption TECHNOLOGICAL FORECASTING AND SOCIAL CHANGE Boyd D; Crawford K Critical questions for big data - Provocations for a cultural, technological, and scholarly phenomenon INFORMACIOS TARSADALOM
  73. 73. Cluster 2: Informetrics/Scientometrics AU TI- SO- Park HW; Leydesdorff L Decomposing social and semantic networks in emerging "big data" research JOURNAL OF INFORMETRICS Park HW An interview with Loet Leydesdorff: the past, present, and future of the triple helix in the age of big data SCIENTOMETRICS Skoric MM The implications of big data for developing and transitional economies: Extending the Triple Helix? SCIENTOMETRICS Fairfield J; Shtein H Big Data, Big Problems: Emerging Issues in the Ethics of Data Science and Journalism JOURNAL OF MASS MEDIA ETHICS Uprichard E Being stuck in (live) time: the sticky sociological imagination SOCIOLOGICAL REVIEW
  74. 74. Cluster 3: Big Data and the Media AU TI- SO- Bruns A; Highfield T; Burgess J The Arab Spring and Social Media Audiences: English and Arabic Twitter Users and Their Networks AMERICAN BEHAVIORAL SCIENTIST Lewis SC; Zamith R; Hermida A Content Analysis in an Era of Big Data: A Hybrid Approach to Computational and Manual Methods JOURNAL OF BROADCASTING & ELECTRONIC MEDIA Mahrt M; Scharkow M The Value of Big Data in Digital Media Research JOURNAL OF BROADCASTING & ELECTRONIC MEDIA Procter R; Vis F; Voss A Reading the riots on Twitter: methodological innovation for the analysis of big data INTERNATIONAL JOURNAL OF SOCIAL RESEARCH METHODOLOGY
  75. 75. Cluster 4: Big Data as a Driver of Change: ‘Challenges and IT Solutions’ AU TI- SO- Rust RT; Huang MH The Service Revolution and the Transformation of Marketing Science MARKETING SCIENCE Leeflang PSH; Verhoef PC; Dahlstrom P; Freundt T Challenges and solutions for marketing in a digital era EUROPEAN MANAGEMENT JOURNAL Hilbert M What Is the Content of the World's Technologically Mediated Information and Communication Capacity: How Much Text, Image, Audio, and Video? INFORMATION SOCIETY Huang MH; Rust RT IT-Related Service: A Multidisciplinary Perspective JOURNAL OF SERVICE RESEARCH Miller HJ Beyond sharing: cultivating cooperative transportation systems through geographic information science JOURNAL OF TRANSPORT GEOGRAPHY
  76. 76. Cluster 5: Big Data and Geography AU TI- SO- DeLyser D; Sui D Crossing the qualitative-quantitative divide II: Inventive approaches to big data, mobile methods, and rhythmanalysis PROGRESS IN HUMAN GEOGRAPHY Wright DJ Theory and application in a post-GISystems world INTERNATIONAL JOURNAL OF GEOGRAPHICAL INFORMATION SCIENCE Crampton JW; Graham M; Poorthuis A; Shelton T; Stephens M; Wilson MW; Zook M Beyond the geotag: situating 'big data' and leveraging the potential of the geoweb CARTOGRAPHY AND GEOGRAPHIC INFORMATION SCIENCE Wilson MW Geospatial technologies in the location-aware future JOURNAL OF TRANSPORT GEOGRAPHY Longley PA Geodemographics and the practices of geographic information science INTERNATIONAL JOURNAL OF GEOGRAPHICAL INFORMATION SCIENCE Shah NH; Tenenbaum JD The coming age of data-driven medicine: translational bioinformatics' next frontier JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION Kwon O; Sim JM Effects of data set features on the performances of classification algorithms EXPERT SYSTEMS WITH APPLICATIONS
  77. 77. Cluster 6: Big Data in the cloud: Information systems related contributions AU TI- SO- Tien JM Big Data: Unleashing information JOURNAL OF SYSTEMS SCIENCE AND SYSTEMS ENGINEERING Miller HE Big-data in cloud computing: a taxonomy of risks INFORMATION RESEARCH-AN INTERNATIONAL ELECTRONIC JOURNAL Lee MY; Lee AS; Sohn SY Behavior scoring model for coalition loyalty programs by using summary variables of transaction data EXPERT SYSTEMS WITH APPLICATIONS Waller MA; Fawcett SE Data Science, Predictive Analytics, and Big Data: A Revolution That Will Transform Supply Chain Design and Management JOURNAL OF BUSINESS LOGISTICS Lycett M 'Datafication': making sense of (big) data in a complex world EUROPEAN JOURNAL OF INFORMATION SYSTEMS Kim C; Lev B Enterprise Analytics: Optimize Performance, Process, and Decisions Through Big Data INTERFACES Lee CH; Chien TF Leveraging microblogging big data with a modified density-based clustering approach for event awareness and topic ranking JOURNAL OF INFORMATION SCIENCE Tien JM The next industrial revolution: Integrated services and goods JOURNAL OF SYSTEMS SCIENCE AND SYSTEMS ENGINEERING Sahoo SS; Jayapandian C; Garg G; Kaffashi F; Chung S; Bozorgi A; Chen CH; Loparo K; Lhatoo SD; Zhang GQ Heart beats in the cloud: distributed analysis of electrophysiological 'Big Data' using cloud computing for epilepsy clinical research JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION
  78. 78. Cluster 7: Techniques to analyse Big Data AU TI- SO- Janowicz K Observation-Driven Geo-Ontology Engineering TRANSACTIONS IN GIS Chen HC; Chiang RHL; Storey VC BUSINESS INTELLIGENCE AND ANALYTICS: FROM BIG DATA TO BIG IMPACT MIS QUARTERLY Wiedemann G Opening up to Big Data: Computer-Assisted Analysis of Textual Data in Social Sciences HISTORICAL SOCIAL RESEARCH-HISTORISCHE SOZIALFORSCHUNG Videla-Cavieres IF; Rios SA Extending market basket analysis with graph mining techniques: A real case EXPERT SYSTEMS WITH APPLICATIONS Prathap G Big data and false discovery: analyses of bibliometric indicators from large data sets SCIENTOMETRICS McKenzie G; Janowicz K; Adams B A weighted multi-attribute method for matching user-generated Points of Interest CARTOGRAPHY AND GEOGRAPHIC INFORMATION SCIENCE Gao S; Liu Y; Wang YL; Ma XJ Discovering Spatial Interaction Communities from Mobile Phone Data TRANSACTIONS IN GIS
  79. 79. Cluster 8: Big Data and Big Brother: Cyber Surveillance AU TI- SO- Hu M Biometric ID Cybersurveillance INDIANA LAW JOURNAL Martinez MG; Walton B Crowdsourcing: the potential of online communities as a tool for data analysis OPEN INNOVATION IN THE FOOD AND BEVERAGE INDUSTRY Sui D Opportunities and Impediments for Open GIS TRANSACTIONS IN GIS Krasmann S; Kuhne S Big Data and Big Brother - what if they met? On a neglected political dimension of technologies of control and surveillance in the research on acceptance KRIMINOLOGISCHES JOURNAL
  80. 80. Cluster 9: Big Data and Decision Support Systems AU TI- SO- Demirkan H; Delen D Leveraging the capabilities of service-oriented decision support systems: Putting analytics and big data in cloud DECISION SUPPORT SYSTEMS Cogean DI; Fotache M; Greavu- Serban V NOSQL IN HIGHER EDUCATION. A CASE STUDY INTERNATIONAL CONFERENCE ON INFORMATICS IN ECONOMY Li T; Kauffman RJ Adaptive learning in service operations DECISION SUPPORT SYSTEMS Julian CD Do Relational Databases Finally Have a Real Competitor? The Struggle of a New Breed - NoSQL INNOVATION AND SUSTAINABLE COMPETITIVE ADVANTAGE: FROM REGIONAL DEVELOPMENT TO WORLD ECONOMIES, VOLS 1-5 Walker S Big Data: A Revolution That Will Transform How We Live, Work, and Think INTERNATIONAL JOURNAL OF ADVERTISING Lovric M; Li T; Vervest P Sustainable revenue management: A smart card enabled agent- based modeling approach DECISION SUPPORT SYSTEMS
  81. 81. Outlook • New field, little work linking the various themes: • BIG DATA the one key denominator • Emerging differentiation • Identified 8-9 clusters in SS/LIS ‘big data’ literature in WoS • The Triple Helix and Big Data • Plenty of space to leave a mark • Very little ground covered • Leydesdorff and Park notable exceptions • Opportunities: • TH occurs implicitly in most social science papers • More conceptual work necessary
  82. 82. SPEECH ACTS IN TELEVISED PRESIDENTIAL DEBATES AND FACEBOOK MESSAGES: THE CASE OF THE 2012 SOUTH KOREAN PRESIDENTIAL ELECTION
  83. 83. Purpose of the current study  With the advent of social networking sites (SNSs), ordinary individuals have opportunities to participate in communication on televised social events and issues.  The present study bridges theories of speech acts and political representation  How did leading and trailing presidential candidates incorporate speech acts into their rhetorical strategies in three consecutive presidential debates during the 2012 presidential election in Korea?  How did their supporters employ speech acts when leaving messages on Facebook fanpages?
  84. 84. Speech acts  Language use goes beyond the boundary of the syntactic structure and its semantic meaning  Language is used to perform speech acts for certain functions such as promising, asking, ordering, and requesting, among others (Austin, 1976; Habermas,1981; Searle, 1969; Wittgenstein, 2009).  Every speech act has three components (Austin; Searle)  A locutionary component (a propositional content component),  An illocutionary component (an action component),  A perlocutionary effect (a consequence of saying something).
  85. 85. Televised presidential debates and speech acts  A few studies have attempted to understand how debate participants use different argumentative styles, linguistic devices, and speech acts.  Lee and Benoit (2005) reported that during the 2002 Korean presidential debates, the candidates used acclaims (52%) more often than attacks (37%) and defenses (11%).  Benoit (2007) reviewed political debates in various countries and concluded that presidential candidates most frequently used acclaims, followed by attacks and defenses.  Bilmes (1992) analyzed the 1992 U.S. vice presidential debate and found that, in addition to assertions, questions were frequently addressed by the candidates.
  86. 86. Televised presidential debates and speech acts  The use of interrogatives can be perceived as an aggressive tactic used by trailing candidates attempting to raise the public's suspicion about the leading candidate's credibility, integrity, morality, and expertise, among others (Wilson & Speder, 1988).  The candidates frequently and strategically asked questions to one another to identify controversial issues and raise the listener's suspicion about the opponent's normative base (Bilmes, 1999).
  87. 87. Televised presidential debates and speech acts  The presidential candidates during the 2004 U.S. presidential debates frequently offered promises and that their verbs included "promise," "swear," and "want" (Marietta, 2009).  Al-Bantany (2013) analyzed a gubernatorial debate and found guarantees and promises to be two most frequently employed commissive speech acts.  Edelsky and Adams (1990), who examined six mixed- gender state and local debates and verified stereotypical differences in communication styles between male and female candidates.
  88. 88. Suggested hypotheses (part one)  H1. Presidential candidates are more likely to use constatives than any other type of speech act.  RQ1. Other than constatives, how frequently do presidential candidates use various types of speech acts during presidential debates?  H2. The trailing candidate is more likely to use directives and interrogatives than the leading candidate during a presidential debate.  H3. The leading candidate is more likely to use commissives than the trailing candidate during presidential debates.  H4. Female candidates are more likely to use expressives than male candidates during presidential debates.
  89. 89. Speech acts on candidates’ Facebook fanpages  With respect to CMC messages, assertives are the dominant type of speech act, followed by expressives and commissives (Hassel & Christensen, 1996; Nastri et al., 2006) .  With respect to SNS messages, expressives are the most widely employed type of speech act, followed assertives, directives, and commissives, claiming that SNS users try to present themselves through the use of humor (Carr et al, 2009; 2012; Ellison, Steinfeild, & Lampe, 2011; Ilyas & Khushi, 2012; Thelwall & Buckley, 2013).  Supporters of leading and trailing candidates may be inclined to use different types of speech acts to actualize the possibility of winning the presidential election.
  90. 90. Suggested hypotheses (part two)  H5. Visitors to presidential candidates’ Facebook pages are more likely to use assertives than any other type of speech act, followed by expressives.  H6. Moon’s Facebook page visitors are more likely to use constatives than Park’s visitors.  H7. Moon’s Facebook page visitors are more likely to use directives than Park’s visitors  H8. Moon’s Facebook page visitors are more likely to use commissives than Park’s visitors.  H9. Moon’s Facebook page visitors are more likely to use quotations than Park’s visitors.
  91. 91. Method  Samples  the debate script was extracted for each candidate from http://www.debates.go.kr: 609 sentences for Park and 776 sentences for Moon  Facebook messages posted on these pages were extracted from December 4, 2012, to December 17, 2012.  Postings were divided based on the debate schedule: six time periods.  A total of 300 messages were randomly selected for each time period for each candidate’s Facebook page.  If there were fewer than 300 messages during a certain period, then all messages were included.
  92. 92. Method  Coding Code Examples Constatives “She doesn’t have any idea about economic democratization,” “He was t oo gentle,” “He definitely won the debate,” and “Mr. Lee. Without natio nal security we can’t achieve welfare either.” Directives “You have to be more aggressive next time,” “Just ignore his stupid accu sation,” “Do not post this kind of stupid comment,” and “Tell me what y our opinion is on the half-tuition policy.” Commissives “I’ll definitely vote in this election,” “We should vote for change,” and “ Let’s vote and end this absurdity.” Expressives “I was so impressed^^,” “Fighting!” “I love all Korean mothers ^^~~^^♥ ♥♥♥♥.” Interrogatives “Do you agree with me?” and “I want to ask how you feel about those pe ople who suffered under your father’s reign.” Quotations* “Lee is giving a speech for Moon http://news1.kr/articles/917472.” Expectatives “If you graduate from a university, I hope our country will be a livable pl ace” and “I want to see president Moon.” *Only quotations were applied to analyze Facebook messages.
  93. 93. Results Speech acts Frequency Percentage Constatives 933 67.4 Directives 35 2.5 Commissives 198 14.3 Expressives 53 3.8 Interrogatives 161 11.6 Expectatives 5 .3 Two candidates’ speech acts during three presidential debates Speech acts Frequency (%) Chi-square P Park Moon Constatives 380 551 2.73 <.05 Directives 11 28 3.72 <.05 Commissives 129 68 46.88 <.01 Expressives 31 22 5.17 <.05 Interrogatives 46 114 17.83 <.01 Expectatives 2 3 .02 n.s. Differences in speech acts between Park and Moon during presidential debates
  94. 94. Results Speech acts Frequency (%) Chi-square P Park Moon Constatives 623 583 .09 n.s. Directives 113 164 9.92 <.01 Commissives 4 44 41.32 <.01 Expressives 521 413 25.09 <.01 Interrogatives 55 55 .01 n.s. Quotations 73 156 32.29 <.01 Expectatives 41 53 2.67 n.s. Total 1430 1468 Differences in speech acts of Facebook visitors between Park and Moon
  95. 95. Results  Both candidates uttered more acclaims than any other speech acts, consistent with the findings of previous research.  The leading candidate used more commissives, whereas the trailing candidate, more aggressive speech acts such as constatives, directives, and interrogatives.  Moon was aggressive in that he used more directives and interrogatives than Park. On the contrary, Park used more commissives and expressives than Moon.
  96. 96. Results  Moon’s fanpage visitors used more commissives and directives than Park’s visitors.  Moon’s visitors used more quotations than Park’s.  Park’s visitors used more expressives than Moon’s.
  97. 97. Concluding remarks  First, the candidates were most likely to employ clams for truth (constatives), promises for the future (commissives), revelations of subjective feelings (expressives), attacks for regulating interpersonal relationships (directives and interrogatives), and expectatives, in that order.  Second, Moon was more likely to attack than Park, and Park was more likely to promise than Moon.  Third, Moon’s Facebook page visitors engaged in interactive relationships with others by using more directives and commissives than Park’s visitors.
  98. 98. Introducing the Oxford Internet Institute (OII) Prof. Ralph Schroeder
  99. 99. • Social sciences department at the University of Oxford • Undertaking rigorous multi- disciplinary research and teaching on the societal impact of the Internet and ICTs (e.g. law, economics, politics & sociology) • Developing methodologically innovative tools and techniques • Training the next generation of Internet-literate researchers. Since our inception we have sought to inform and shape policy and practice.
  100. 100. Taught Courses • 50+ graduate students from wide variety of disciplinary backgrounds, and from industry or government • DPhil Information, Communication and the Social Sciences: supports single or multi-disciplinary research. • MSc in Social Science of the Internet: 1 year Masters delivering core training in social science methods and statistics, understanding of the Internet’s technical architecture and regulatory framework, social dynamics of Internet’s impact, in-depth disciplinary study e.g. Internet Economics or Law plus cutting edge tools for digital social research. • Annual Summer Doctoral Programme (2 weeks) for advanced PhD students completing Internet-related theses across a variety of disciplines.
  101. 101. Michaelmas Hilary Trinity Methods Social Research Methods and the Internet Part I Social Research Methods and the Internet Part II Core Survey courses Social Dynamics of the Internet Internet Technologies and Regulations Options Two Option Courses Dissertation Dissertation
  102. 102. Two Options • Digital Era Government and Politics • Internet Economics • Law and the Internet • Online Social Networks • Learning, the Internet and Society • Big Data and Society • Subversive Technologies • ICTs and Development • Digital Social Research
  103. 103. OII Research • Topics covered across Governance and Democracy, Everyday Life, Science & Learning, Network Economy, Shaping the Internet • Social science faculty with computer science skills • Making major contributions to social science, e.g. addressing the challenge of Big Data • Field-leading methodological innovation e.g. Facebook & NameGenWeb, OxLab. • Biennial benchmarking and analysis of UK Internet use and non-use (OxIS) • Compelling presentation of data and findings to maximise public engagement (e.g. iBook, Visualising Data).
  104. 104. Other relevant projects • Future Home Networks & Services (Ian Brown & Joss Wright): researching and developing security frameworks for sharing between networks and devices, and cloud services; • Oxford e-Social Science Project (Ralph Schroeder & Eric Meyer): aims to understand how e-Research projects negotiate various social, ethical, legal and organizational forces and constraints; • The Learning Companion Project (Rebecca Eynon & Yorick Wilks): evaluates the feasibility of a computer-based digital tool to help adults whose engagement with learning is tentative make productive use of the Internet for learning projects. • Privacy Value Networks (Ian Brown): producing an empirical base for developing concepts of privacy across contexts and timeframes, addressing a current lack of clarity of what privacy is and what it means to stakeholders in different usage scenarios
  105. 105. Research Examples • People and Research • Big Data: UK Government • OxIS • Political Science: Helen Margetts • Geography: Mark Graham • Social Network Analysis: Bernie Hogan • Oxford e-Social Science Project: Dutton, Schroeder, Meyer
  106. 106. Big Data: UK Government Online . • JISC UK Web Domain Dataset (30 Tb) of .uk ccTLD from 1996-2010 • Here shows link structure of government (.gov.uk) in 2012 • Data can reveal change in government relationships and structure over time
  107. 107.  Data  Internet Archives data of .uk back to 1996  Annual crawls of .uk websites since 2013  2.7 billion nodes, 40TB compressed  Features  Full text search (in progress, IHR)  Network analysis (OII)  N-gram analysis  Limitations  Page content data access limited
  108. 108. Growth of subdomains N.B. y-axis on log scale
  109. 109. Relative sector size on the web
  110. 110. Sectoral linking 2010
  111. 111. OII Faculty
  112. 112. Use by Age (QH14 by QD1) OxIS 2005: N=2,185; OxIS 2007: N=2,350; OxIS 2009: N=2,013 16
  113. 113. Which is more Important: Age or Income? Internet Users in Each Age-Income Category (percents) Age Groups Income 14-44 45-64 65+ Up to £20K/year 71.3 39.3 21.3 £20-40K/year 92.6 78.3 49.0 Over £40K/year 97.0 96.4 75.0 • OxIS 2009: N=1,318 Internet Users
  114. 114. Use by Education (QH14 by QD14) OxIS 2007: N=2,350; OxIS 2009: N=2,013 (Basic: N=901; Further: N=510; Higher: N=360). Note: Students were excluded. 18
  115. 115. Web 2.0 User Creativity & Production Online (QC10 and QC31) Current users. OxIS 2005: N=1,309; OxIS 2007: N=1,578; OxIS 2009: N=1,401 Note. Social networking question changed in 2009. 19
  116. 116. Helen Margetts ESRC Professorial Fellowship 2011-2014 The Internet, Political Science And Public Policy Re-examining Collective Action, Governance and Citizen-government Interactions in the Digital Era • Using the internet to generate ‘real’ transactional data about political behaviour (including webmetrics, datamining and experiments) 8,327 petitions scraped from No 10 Downing Street site, all new ones 2009-2010 95% of petitions fail to reach 500 (number necessary for official reply) Number of signatures on launch day crucial to whether it reaches 500
  117. 117. •Social network map of Bernie Hogan’s FB ties, Dec. 2008; •Proof of concept network that led to creation of NameGenWeb Mapping Personal Networks
  118. 118. Family Local Friends Three co-worker groups Friends
  119. 119. Mark Graham: Total number of Wikipedia articles per 100,000 people
  120. 120. •Mark Graham & Bernie Hogan’s project investigates inequalities in the creation of knowledge. • Map reveals uneven spread of geo-tagged Wikipedia articles 2011-12.
  121. 121. Sandra Gonzalez-Bailon USENET Political Discussions (1999-2005) 0 2 4 6 8 x10000 09/1999 09/2000 09/2001 09/2002 09/2003 09/2004 gun whiteblack newswar people hateworld partyfree deathgood mancrime housetime moneyboy abortion flag 0:1 white gun news people war black time house party world goodcut death power hateman fraudfree truthcrime 0:1 war white worldgun terrorist newstime people housegood deathhate mandeadblack peacetruthfree lettergod 0:1 warworld news people whitegood time peace gun death hate house dead black terrorist party f ree man truth lie 0:1 war news white worldtime peoplegood hatedead manhouse partydeath freeblack lie truthguntorture terrorist 0:1 war news timeworld people hatewhite socialdead goodman houseparty goddeath fraudwinfree gunblack
  122. 122. Emotions and Public Opinion
  123. 123. Oxford e-Social Science Project • Social shaping and implications of e-Research • Collaborative project with: • SBS / InSIS group • OeRC • ESRC: 6 years of funding + multiple follow-on projects
  124. 124. Source: Schroeder, R., Meyer, E.T. (2009). Gauging the Impact of e-Research in the Social Sciences. Paper presented at the 104th American Sociological Association Annual Meeting, August 8-11, San Francisco, California.
  125. 125. Source: Meyer, E.T., Park, H-W., Schroeder, R. (2009). Mapping Global e-Research: Scientometrics and Webometrics. Proceedings of the 5th International Conference on e-Social Science, June 24-26, Cologne, Germany.
  126. 126. Source: Meyer, E.T., Schroeder, R. (2009). Untangling the Web of e-Research: Towards a Sociology of Online Knowledge. Journal of Informetrics 3(3):246-260.
  127. 127. Source: Meyer, E.T., Schroeder, R. (2009). Untangling the Web of e-Research: Towards a Sociology of Online Knowledge. Journal of Informetrics 3(3):246-260
  128. 128. Source: Schroeder, R., Meyer, E.T. (2009). Gauging the Impact of e-Research in the Social Sciences. Paper presented at the 104th American Sociological Association Annual Meeting, August 8-11, San Francisco, California.
  129. 129. For more information see our website: http://www.oii.ox.ac.uk Twitter: @oiioxford
  130. 130. Big Data, Big Brother, and Social Science Ralph Schroeder Collaborators: Eric T. Meyer, Linnet Taylor, Josh Cowls, Greg Taylor, Monica Bulger Asia Triple Helix Society, Daegu, 25th June, 2014
  131. 131. Overview • Projects • Questions • Issues • Definition • How knowledge advances • Examples • Big Data Issues in Research and Beyond • Policy Implications • Conclusion
  132. 132. Accessing and Using Big Data to Advance Social Science Knowledge • Funded by Sloan Foundation • Data sources • 100+ interviews, mainly with social scientists • Reports, workshops • Publications, conferences • No representative sample, but some patterns of disciplinary and skills background and career trajectory
  133. 133. See http://www.oii.ox.ac.uk/research/projects/?id=98
  134. 134. Data-driven economic models: challenges and opportunities of big data • Funded by Research Councils UK (RCUK), New Economic Models in the Digital Economy (NEMODE) network • Data Sources: – 25+ interviews – Case studies – Issues include how models relate to national contexts (ie. privacy laws in Germany), where skills are located (plus gaps), use of public/private data, standardization
  135. 135. Source: http://www.forbes.com/sites/davefeinleib/2012/06/19/the-big-data-landscape/
  136. 136. Source: Leonard John Matthews, CC-BY-SA (http://www.flickr.com/photos/mythoto/3033590171)
  137. 137. Spurious Correlations
  138. 138. Twitter-bots OII master’s students Alexander Furnas and Devin Gaffney saw a large spike in then-US presidential candidate Mitt Romney’sTwitter followers, and decided to look at the new followers: Furnas, A. and Gaffney, D. (2012). ‘Statistical Probability That Mitt Romney's New Twitter Followers Are Just Normal Users: 0%’. The Atlantic, July 31, http://www.theatlantic.com/technology/archive/2012/07/statistical-probability-that-mitt-romneys-new-twitter-followers-are-just-normal-users-0/260539/ (accessed August 31, 2012).
  139. 139. Google Images: Big Data
  140. 140. Source: Hill, K. (Feb 16, 2012). Forbes.com. Available at: http://www.forbes.com/sites/kashmirhill/2012/02/16/how-target-figured- out-a-teen-girl-was-pregnant-before-her-father-did/ Based on Duhigg, C. (Feb 16, 2012). “How Companies Learn Your Secrets.” New York Times Magazine.
  141. 141. 113 240 278 367 558 1,195 1,538 2,350 3,960 6,787 7,276 9,010 - 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000 10,000 1st Q 2nd Q 3rd Q 4th Q 1st Q 2nd Q 3rd Q 4th Q 1st Q 2nd Q 3rd Q 4th Q 2010 (n=998) 2011 (n=5,641) 2012 (n=27,033) Number of News Articles on Big Data Source: Nexis data compiled by Meyer & Schroeder
  142. 142. Big data in the commercial world • Commercial uses are: ‘in house’, ‘outsourced own data’, ‘data analysis as a consultancy service’ • Careers in data analysis entail as a baseline computer science/statistical expertise, plus different domains of ‘sorting people’ and being able to ‘manipulate’ them (ie. predict their behaviour)
  143. 143. Definition • ‘Big data’ – the advance of knowledge via a leap in the scale and scope in relation to a given object or phenomenon ‘Data’ – Belongs to the object – ‘taking…before interpreting’ (Ian Hacking) • the view that ‘all data are of their nature interpreted’ is misleading: ‘data are made, but as a good first approximation, the making and taking come before interpreting’ – The most atomizable useful unit of analysis
  144. 144. Computational Manipulability? • ‘the distinctiveness of the network of mathematical practitioners is that they focus their attention on the pure, contentless form of human communicative operations: on the gestures of marking items as equivalent and of ordering them in series, and on the higher-order operations which reflexively investigate the combinations of such operations’ • ‘mathematical rapid-discovery science…the lineage of techniques for manipulating formal symbols representing classes of communicative operations’ • Why is big data a big deal? Manipulability, plus new data sources
  145. 145. Research computing The Grid Supercomputing Clouds Big Data Web 2.0
  146. 146. Digital Objects and their Referents Digital Object (Examples: Twitter, Tesco Loyalty card information Real World (People / Physical Objects) Represent / Manipulate
  147. 147. Representing Manipulating Limits Digital Data
  148. 148. 010101 Knowledge
  149. 149. Uses and Limits • Big data research uses (academic, commercial, government) are limited to the exploitation of suitable objects, and the objects which ‘give off’ digital data, and the phenomena they lay bare, are limited • The knowledge produced is aimed at ‘sorting people’ and advancing ‘representing and intervening’ (but without ‘manipulating’, except where this is warranted by practical economic and political objectives) • Difference commercial versus academic world is that knowledge provides competitive and practical advantage as against advancing (high-consensus rapid-discovery) knowledge – The limits in both cases are the objects (to which the data ‘belong’), and that need to have available digitally manipulable data points • How available these objects are differs, but also… – Causation and theoretical embedding matters for academic social science – For commercial (and non-academic uses), ‘predicting’ consumer choices and other behaviours, for limited purposes and without increasing scientific knowledge, is good enough • There are many objects, for non-academics and scientists to humanities scholars (physical, human, cultural), but they are not infinite • This availability, not skills or other issues, determines the future of big data research
  150. 150. 113 240 278 367 558 1,195 1,538 2,350 3,960 6,787 7,276 9,010 - 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000 10,000 1st Q 2nd Q 3rd Q 4th Q 1st Q 2nd Q 3rd Q 4th Q 1st Q 2nd Q 3rd Q 4th Q 2010 (n=998) 2011 (n=5,641) 2012 (n=27,033) Number of News Articles on Big Data Source: Nexis data compiled by Meyer & Schroeder
  151. 151. Platform Paper Size of Data in relation to phenomenon investigated Theoretical question/practical aim Key findings Facebook Backstrom et al. (2012) 69 billion friendship links between 721 million Facebook users Re-examine Milgram’s ‘six degrees of separation’ online Four degrees of separation on Facebook Ugander et al. (2012) 54 million invitation emails to Facebook users How does structure of contacts affect invitation acceptance? Not number of contacts, but number of distinct contexts, matters for acceptance Bond et al. (2012) 600000 Facebook users Facebook experiment about how to mobilize voters Voters can be mobilized via Facebook friends more than via informational messages Twitter Kwak et al. (2010) 1.47 billion directed Twitter relations Is Twitter a broadcast medium or a social network? Most use is for information, not as a social network Cha et al. (2010) 1.7 billion tweets among 54 million users Who influences whom? Top influentials dominate, but some variation by topic Bakshy et al. (2011) 1.6 million Twitter users Who influences whom? ‘Ordinary user’ influencers can sometimes be more effective than top influencers Wikipedia Loubser (2009) All Wikipedia activity How is editing organized? Administrators can impact negatively on participation Yasseri, Kertesz (2012) Editorial activity on Wikipedia, especially reverts Understanding conflict and collaboration Types of conflicts can be modelled West, Weber and Castillo (2012) Wikipedia contributions related to Yahoo! browsing What characterizes Wikipedia contributors’ information behaviour compared to Wikipedia readers and non-readers Wikipedia contributors are more ‘information hungry’, especially about their topics
  152. 152. Example 1: Search engine behaviour Waller’s analysis ofAustralian Google Users Key findings: - Mainly leisure - > 2% contemporary issues - No perceptible ‘class’ differences Novel advance: - Unprecedented insight into what people search for Challenge: - Replicability - Securing access to commercial data
  153. 153. ? ? ? ? ? ? ? ? ? “Surprisingly, the distribution of types of search query did not vary significantly across the different Lifestyle Groups (p>0.01).” Source: Waller, V. (2011). “Not Just Information:Who Searches for What on the Search Engine Google?” Journal of the American Society for Information Science & Technology 62(4): 761-775.
  154. 154. Example 2: Large-scale text analysis Michel et al. ‘culturomic’ analysis of 5 Million Digitized Google Books and Heuser & Le-Khac of 2779 19th Century British Novels Key findings: - Patterns of key terms - Industrialization tied to shift from abstract to concrete words Novel advance: - Replicability, extension to other areas, systematic analysis of cultural materials Challenge: - Data quality
  155. 155. J Michel et al. Science 2011;331:176-182
  156. 156. Example 3: Social network or news? Kwak et al.’s analysis ofTwitter Key findings: - 1.47 billion social relations - 2/3 of users are not followers or not followed by any of their followings - Celebrities, politicians and news are among top 20 being followed Novel advance: -Volume of relations and topics Challenge: - News or social network needs to be contextualized in media ecology - Securing access to commercial data
  157. 157. (Big) data definition enables pinpointing impacts and threats • ‘Google Plus may not be much of a competitor to Facebook as a social network, but…some analysts…say that Google understands more about people’s social activity than Facebook does.’ – New York Times, 15.2. 2014, p. A1 ‘The Plus in Google Plus? It’s Mostly for Google’. • Facebook Likes: ‘Predicting users’ individual attributes and preferences can beused to improve numerous products and services. For instance, digital systems and devices (such as online stores or cars) could be designed to adjust their behavior to best fit each user’s inferred profile…online insurance…advertisements might emphasize security when facing emotionally unstable (neurotic) users but stress potential threats when dealing with emotionally stable ones’ – ‘Private traits and attributes are predictable from digital records of human behavior.’ Kosinski M, Stillwell D, Graepel T.,Proc Natl Acad Sci 2013 Apr 9;110(15):5802-5. • More powerful knowledge will enable better services, and more manipulation
  158. 158. ‘Big data‘ for understanding society • Real-time transactional data (unlike survey data, traditional staple of social science) • Outside capability of normal desktop computing environment (‘Too big to handle’) • Big potential for understanding institutions and individual behaviour
  159. 159. Social Science and Big Data Research • Dominated by social media • Issues of ‘whole universe’ – What population, offline and online, does it represent – Data quality and replicability – How does ‘modality’ determine findings about implications • How to embed the research – In existing theory (but also advance theory) – In existing ecology of media uses in society (including ones that extend existing ones)
  160. 160. Scientificity and Big Data: Pro and Con • Pro – Replicability, extension to new domain – ‘Total’ datasets, ‘whole universe’ – (Often) no sampling needed, data for all behaviour and over whole existence – Ready made manipulability – Powerful relation of data to object • Con – Limited access to object, skills needed for manipulability – (Often) not known who users are – No or little knowledge of how (commercial) data were gathered – Researcher does not ask what is of interest without ‘givenness’ – Datasets capture limited dimensions, and about one object – Object in isolation, not framed for social change significance
  161. 161. Ethical and Social Issues in Big Data Research • Objects with ‘total’ knowledge (universes) – Danger is inferring behaviour not of individuals, but of classes of people • Asymmetry of knower and the subjects of knowledge is greater than elsewhere • Based not on individuals’ but on aggregate behaviour – Hence only utilitarian, not Kantian justification? • Why does prediction or uncovering laws of behaviour ‘grate’? • Benefits: greater scientific power and more specific details • Relation to smaller data? ‘Creep’ • Solution: ethical = greater researcher and public awareness, regulatory (would apply to academic researchers?) = prevent legal and specific harms
  162. 162. Other positions on Big Data Implications 1 • Mayer-Schoenberger and Cukier, boyd and Crawford argue that not all information can or should be captured – No, need to create the legal and ethical social space which protects the individual. The solution does not rely on denying the powerfulness of knowledge, but harnessing it appropriately. • Mayer-Schoenberger and Cukier solution of 1.more transparent algorithm, 2. Certifiying validity of algorithm 3. Allowing disprovability of prediction (p.176) – – Yes, but within social science, solution is to make knowledge more scientific. • Underlying all these problems is more powerful knowledge – This goes against free, untrammelled behaviour – Solution: Society becomes more self-aware and shapes knowledge to constrain it • Crawford, Marwick: big data is product of neoliberal capitalism? No, uses by different societies, and for purposes apart from ‘neoliberal capitalist’ ones, such as open government data and Wikipedia analysis
  163. 163. Other Positions on Big Data Implications 2 • Savage and Burrows: ask are commercial data outpacing social science? • Boyd and Crawford: does big data raise epistemological conundrums, and isn’t it always already (social) contextual ? • Mayer-Schoenberger and Cukier: what are the political and commercial harms of wrong knowledge, especially when it changes ‘everything’? ... No ... • Knowledge depends on the relation between research technologies and the advance of knowledge • The threats and opportunities are not contextual, but depend on how more powerful knowledge is used • Big data contributes to more ‘scientific’ (i.e. cumulative) social sciences, but within limits, and there are limits to commercial and political uses too
  164. 164. Consumer (and gov’t) Big Data • Consumer data and privacy (ie. Target pregnancy case) – Solution: data protection • Consumer data and prediction and control (ie. click behaviour): affects consumer without transparency, predictive privacy harm – Solution: transparency, ‘due process’ (Crawford and Schultz) • Consumer data – and government data - and exclusion from benefits thereof (ie. no or little use of digital devices) - if not captured by data, left out – Solution: Data antisubordination (Lerman) – Solution: government may need more data about us (and counteract the data invisibility of parts of the population) • Consumer data from digital media (ie. search engines) – manipulate what is found without transparenyc, inappropriate personalization (Pariser) – Solution: transparency, consumer protection
  165. 165. Big Data and Policy • Probabilistic rather than ‘causal’ commercial and government uses of data (ie. profiling) - only probable, not definite causal behaviour of data emitters established (Mayer-Schoenberger and Cukier) – Solution: more accurate knowledge • Exposure of Data emitter because of identifiers in large- scale and linked data (Netflix, AOL, Google Streetview, National Security Administration), such that anonymization does not work – Solution: data protection, better anonymization, opting out, consent • Social media used in authoritarian regimes for control (Weibo in China) – Solution: more commercial independence, more civil society pushback, researcher non-cooperation
  166. 166. Future of Big Data Research • Difference commercial versus academic world is that knowledge provides competitive advantage as against advancing (high-consensus rapid-discovery) knowledge • The limits in both cases are the objects (to which the data ‘belong’), and that need to have available digitally manipulable data points • How available these objects are differs • There are many objects, for non-academics and scientists to humanities scholars (physical, human, cultural), but they are not infinite • This availability, not skills or other issues, determines the future of big data research • A Golden Age of Quantification and New Sources of Data…A Dark Age (so far) of understanding new online phenomena and their social significance
  167. 167. Outlook and Implications • There is an overlap between real world research and the world of academic research which is closer than elsewhere – because this is the research front in both – because they share common objects • For research – Develop theoretical frame in which to embed big data (for social media), including power/function, relation to traditional media, and role in society • For society – Awareness of how research can generate transparency and manipulability • Big Brother? – Yes, but also Brave New World of Omniscience, with Social Science as Handmaiden
  168. 168. Additional readings and references Bond, Robert et al. (2012). ‘A 61-million-person experiment in social influence and political mobilization’, Nature 489: 295–298. Bruns, A. and Liang,Y.E. (2012). ‘Tools and methods for capturingTwitter data during natural disasters’, First Monday, 17 (4 – 2), http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/viewArticle/3937/3193 Furnas, A. and Gaffney, D. (2012). ‘Statistical ProbabilityThat Mitt Romney's NewTwitter Followers Are Just Normal Users: 0%’. The Atlantic, July 31, http://www.theatlantic.com/technology/archive/2012/07/statistical- probability-that-mitt-romneys-new-twitter-followers-are-just-normal-users-0/260539/ (accessed August 31, 2012). Giles, J. (2012). ‘Making the Links: From E-mails to Social Networks, the DigitalTraces left Life in the ModernWorld areTransforming Social Science’, Nature, 488: 448-50. Kwak, H. et al. (2010). ‘What isTwitter, a Social Network or a News Media?’ Proceedings of the 19th InternationalWorldWide Web (WWW) Conference, April 26-30, 2010, Raleigh NC. Manyika, J. et al. (2011). ‘Big data: the next frontier for innovation, competition and productivity’, McKinsey Global Institute, available at: http://www.mckinsey.com/insights/mgi/research/technology_and_innovation/ big_data_the_next_frontier_for_innovation (last accessed August 29, 2012). Silver, Nate. (2012). The Signal and the Noise:The Art and Science of Prediction. London:Allen Lane. Tancer, B. (2009). Click:What Millions of People are Doing Online andWhy It Matters. NewYork: Harper Collins, 2009. Wu, S. , J.M. Hofman,W.A. Mason, and D.J. Watts, (2011). ‘Who says what to whom on twitter’, Proceedings of the 20th international conference onWorld WideWeb. (on DuncanWatts webpage, http://research.microsoft.com/en-us/people/duncan/, last accessed August 29, 2012).
  169. 169. Project Papers Schroeder, Ralph (Forthcoming). ‘Big Data: Towards a More Scientific Social Science and Humanities’ in Mark Graham and William H Dutton (eds.), Society and the Internet: How Networks of Information are Changing our Lives. Forthcoming. Schroeder, Ralph, & Taylor, Linnet (Forthcoming). ‘Is bigger better? The emergence of big data as a tool for international development policy.’ GeoJournal. Meyer, Eric T., Schroeder, Ralph, & Taylor, Linnet (2013, August). ‘Big Data in the Study of Twitter, Facebook and Wikipedia: On the Uses and Disadvantages of Scientificity for Social Research.’ Paper presented at the proceedings of the Annual Meeting of the American Sociological Association. (being submitted) Schroeder, Ralph, & Taylor, Linnet. ‘Big Data and Wikipedia Research: Social Science Knowledge across Disciplinary Divides’. Submitted to Information, Communication and Society. Taylor, Linnet. ‘No place to hide? The ethics and analytics of tracking mobility using African mobile phone data. Submitted to Population, Space and Place. Meyer, Eric T., Schroeder, Ralph, & Taylor, Linnet. ‘Big Data in the Social Sciences: Towards a New Research Paradigm?’ (being submitted). Meyer, Eric T., Schroeder, Ralph, & Taylor, Linnet (2013, November). ‘The Boundaries of Big Data.’ Paper presented at SIG-SI Symposium, ASIST 2013, November 1-6, 2013, Montreal, Quebec, Canada. Schroeder, Ralph and Cowls, Josh. ‘Answering Questions and Questioning Answers in the Era of Big Data.’ In preparation. Taylor, Linnet, Meyer, Eric T., & Schroeder, Ralph. ‘Bigger and better, or more of the same? Emerging practices and perspectives on big data analysis in economics”. Forthcoming in Big Data & Society. Cowls, Josh. ‘The Crowd in the Cloud?’, forthcoming presentation and IPP 2014’ Cowls, Josh ‘Big Data and Policy Implementation’, in preparation. Schroeder, Ralph ‘Big Data and Policy Implications’, in preparation.
  170. 170. Oxford Internet Institute With support from: Ralph Schroeder ralph.schroeder@oii.ox.ac.uk http://www.oii.ox.ac.uk/people/?id=26 See http://www.oii.ox.ac.uk/research/projects/?id=98
  171. 171. Understanding “Wedge-Driving” Rumors Online during a Political Crisis: Insights from Twitter Analyses during Korean Saber Rattling 2013 K. Hazel Kwon, PhD, ASU C. Chris Bang, MA, Univ. at Buffalo H. R. Rao, PhD, Univ. at Buffalo
  172. 172. Rumors Revisited • Unofficial Information Sharing in Social Media • Unofficial Information = Rumors = Representation of bottom-up, spontaneously shaped public opinions (Knapp, 1944; Peterson & Gist, 1951; Turner & Killian, 1987) • Haven’t been studied much until recently.
  173. 173. Goals of the Study • Theoretically: Understanding social media rumormongering as a contentious process of collectively constructing meaning under a high uncertainty • Methodologically: Demonstrating how semantic network analytic approach can help textual, discourse analysis of rumors.
  174. 174. Public Opinions • Public Opinions: (1) citizen responses as opposed to governing actors; (2) expressed openly instead of privately reserved; (3) relevant to social affairs with a potential influence on political process • In modern political system: Public Opinion = Opinion Polling Results
  175. 175. Opinion Polling… • A top-down, institutionalized construction of public opinions • Quantitative, limited conveyance of opinion patterns • Mainly for social control • Overemphasis on a “rational” process of opinion formations
  176. 176. Rumors: Improvised Public Opinions • Alternative indicators of opinion climate • Bottom-up, unstructured construction of social affairs • A less normative, less rational process of public sense-making: “Affect-laden” • Help qualitative, granular understanding of opinion patterns
  177. 177. Textual Analysis of Rumors • Social Psychology of Rumors • Textual Analysis of Rumors - Only a few studied due to the lack of text data - Advantage of utilizing social media data (i.e. Twitter) for both theoretical and practical reasons
  178. 178. Wedge-Driving (WD) Rumors • 3 rumor types during a crisis: wish, dread, WD • WD rumors: a moniker for unverified propositions toned with derogatory toward a specific target group or individuals representative of the group • Reflective of social structures for emotional contagions; subconscious roots of intergroup conflict; inverse indicator of social capital; prevailing norms and way of thinking
  179. 179. Empirical Research Questions: To what extent does rumoring happen in social media when a society faces a social/political crisis? Do WD rumors reveal distinctive narrative characteristics in comparison to other types of informal public discourses?
  180. 180. Case: Korean Saber Rattling 2013 • Rumormongering = uncertainty (ambiguous situation) x anxiety (issue importance) • Saber Rattling between North and South Koreas 2013 picked up as a proper case to explore social media rumoring [North Korea = NK; South Korea = SK]
  181. 181. Small-Scale Content Analysis • Quota sampling of 2,500 non-redundant, unique tweet messages (2,352 after filtering) from a total of 207,992 tweets collected between Feb 18 and Mar 14, 2013 • 7 search keywords: 북한(North-Korea), 북핵(North-Korea-Nuclear), 북조선(North- Chosun), 핵무기(Nuclear-Weapon), 핵폭탄(Nuclear-Bomb), 핵실험(Nuclear- Experiment), 김정은(Kim-Jung-Un)
  182. 182. Content Analysis • Dummy coding: (1) informational ambiguity (84.5% agreement), (2) propositional statement (88.9% agreement), (3) hostility towards others than NK (and its politicians) • 3 Groups categorized: (1)&(2)&(3) = WD rumor (1)&(2) = General rumors (GR) The rest = Non-rumors (NR)
  183. 183. Semantic Network Analysis • Words selected based on Bonferroni- adjusted z-tests of word frequency comparisons among the 3 groups • Co-occurrence matrix for each group • Degree & Eigenvector centralities • Clauset-Newman-Moor clustering algorithms
  184. 184. General Results • 25% NR message (62 words), 36.4% WD messages (99 words), and 38.6% GR messages (41 words) • Two centrality scores highly correlated: Spearman’s ρ = .991 for NR, .946 for GR, .943 for WD • 4 semantic clusters in NR network; 5 in WD network; 7 in GR network
  185. 185. Non-Rumor Semantic Network
  186. 186. Non-Rumor Semantic Network
  187. 187. Non-Rumor Semantic Network
  188. 188. Non-Rumor Semantic Network
  189. 189. Non-Rumor Semantic Network
  190. 190. NR network highlights… • Formal, top-down responses to the threat, in a broader geopolitical context.  SK’s political and military capability (C1)  Foreign diplomacy of both Koreas (C2&C3)  International responses to the threat (C4)
  191. 191. WD Semantic Network
  192. 192. WD network highlights • Derogatory themes:  Defaming historic or current politicians (C1), even a public figure in a non- political sector (C2)  Distorting a historical event not directly related with the current threat (C3)  Evoke Cold-War rhetoric to attack opposite political beliefs (C4&C5).
  193. 193. GR semantic network
  194. 194. GR network
  195. 195. GR network
  196. 196. GR semantic network
  197. 197. GR network highlights… • Bottom-up reaction to the threat  the public’s curiosity about the NK’s readiness of kinetic warfare (C1&C2) and their true motivations behind threatening (C3).  Trivialization (C2&C5&C6)  Conveyance of hope (C4&C7)
  198. 198. Discussion & Conclusions • Nontrivial portion of spontaneous, less- than-rational public responses to social or political affairs, i.e. in time of crisis: Calls for understanding rumor publics
  199. 199. • Non-rumors: similar to institutional polling (e.g. Gallup questionnaire) • General-rumors: derivative of the news agenda but mutated into the bottom-up desires to cope with fears: In forms of Guesswork, witticism, pipe-dreaming • WD rumors: deviate a lot, mainly ideological contention between pro-peace and pro-constraint political faction, intertwined with collective memory in histories
  200. 200. Limitation & Future Research • Threw away a large amount of available data due to limited methods • Needs to incorporate a machine-learning approach to scale up research
  201. 201. A social network framework to analyze the cultural contents of Kpop across countries Ji-Young Park & Ji-Young Kim (PhD student, YeungNam University) Wayne Weiai Xu (PhD student, State University of New York at Buffalo) Han Woo Park (Professor, Ph.D.)
  202. 202. Contents • Cultural phenomenon of the Korean wave • Variety of Data procedure - Data preparation - Data process • Social network analysis framework - online cultural contents of Kpop
  203. 203. Cultural phenomenon of the Korean wave • Hallyu(한류: Korean Wave) is a neologism referring to the increase in the popularity of South Korean culture since the late 1990s. The term was originally coined in mid-1999 by Beijing journalists who were surprised by China's growing interest for South Korean cultural exports. They subsequently referred to this new phenomenon as "Hánliú" (韓流), which literally means "flow of Korea".
  204. 204. Cultural phenomenon of the Korean wave • Cultural exports such as Hallyu (“Korean Wave”) embody the global influence of local pop culture. • The promotion of strategic cultural offerings can enhance the national image and strengthen the country’s entertainment industry (Maitland & Bauer, 2001). • The global diffusion of cultural offerings has been increasingly facilitated through social media, a phenomenon that has drawn growing scholarly attention in recent years (see Kim, Heo, et al., 2013).
  205. 205. Web 1.0 Korean Wave Web 2.0 Korean Wave Period Early 2000s 2010s Genre Mostly TV dramas Multiple Contents (e.g. K-pop, Online games) Location Asia Region Centered Globalization Users’ main media platform Websites Social Media (e.g, Twitter, youtube) Marketing strategy Top-down (Government) Bottom – up (fans, market players) The Change of the Korean Wave Source : revised from SERI Quarterly, Oct. 2011. Cultural phenomenon of the Korean wave
  206. 206. • This study focuses on Kpop and a Korean rapper Psy’s Gangnam Style (GS)
  207. 207. Research Questions • What is the communication patterns among international fans of Kpop across countries ?
  208. 208. • Various kinds of online data are used in current paper. • The big data-based analysis programs, including the Webometric Analyst 2.0 and Webonaver & Webogoogle, are employed to retrieve and parse data from the World Wide Web • Data collected are moved to SNA tools such as NodeXL, UciNet, Pajek, and ConText for quantitative investigation
  209. 209. • (1) Web documents on Korean singers • (2) Visibility of Korean singers at popular social media sites • (3) Communication patterns among international fans of Kpop across countries Social network analysis framework
  210. 210. Social Network Analysis Framework Data procedure Method SNA tool (1) Web documents on Korean singers - Scrape keyword(Korean singer) hit count in search result - Scrape keyword(Korean singer) title, phrase & url in search result Webometrics Analysis NodeXL, UciNet, Pajek, and ConText (2) Visibility of Korean singers at popular social media sites - Data collect keyword(Korean singer)’s social media activity like Singer`s follower, following, tweets on Twitter Webometrics Analysis (3) Communication patterns among international fans of Kpop across countries - Data collect using Webometrics Analyst 2.0 - video ID, published date, updated date, video title, video url, author name, dislike, likes viewcount, favorite count - recent 1,000 comments - subscription Network Analysis
  211. 211. Social network analysis framework • (1) Web documents on Korean singers - Webonaver, Webogoogle
  212. 212. Social network analysis framework • (1) Web documents on Korean singers - Webonaver as a scrapper tool -NaverScrapper - ScrapperTools related Naver, Search Engine and Portal -*Using OpenAPI on Naver -Scrape keyword hit count in search result -Scrape keyword title, phrase & url in search result -박한우, 박세정, David Stuart, 이승욱(2009). API를 활용한 검색 프로그램 WeboNaver의 이해와 적용: 18대 국회의원 웹 가시성 분석과 신종플루 관련 단 어의 연관성 분석. Journal of the Korean Data Analysis Society. 11권 6호(B). 3427-3440 -It can be download from http://hanpark.net (allow autherized )
  213. 213. Social network analysis framework • (1) Web documents on Korean singers - WeboGoogle as a scrapper tool -WeboGoogle - ScrapperTools related Google, Search Engine -*Using Custom search API on Google -Scrape keyword hit count in search result -Scrape keyword title, phrase & url in search result - Keyword co-occurrence of the sites' domains based on their symmetrical relationships by using Boolean operators.
  214. 214. Social network analysis framework • (1) Web documents on Korean singers - WeboGoogle as a scrapper tool - The results based on a total of 3,320,000 hit counts from Google-indexed web documents for the search query "Gangnam Style“ on August 14, 2012, - indicate 39.0% of all returned web documents from YouTube.com, followed by AllKpop.com (9.0%) and blogs.wsj.com (3.0%).
  215. 215. Social network analysis framework • (2) Visibility of Korean singers at popular social media sites -Twitter, Facebook Using Nodexl, an open-source software tool, to collect and analyze these Tweets (Hansen, Shneiderman & Smith, 2010). Collect keyword(singer)’s social activity like follower, following, tweets.
  216. 216. Social network analysis framework • (3) Communication patterns among international fans of Kpop across countries Webometric Analyst analyses the web impact of documents or web sites and creates network diagrams of collections of web sites, as well as creating networks and time series analysis of social web sites (e.g., YouTube, Twitter) and some specialist web sites (e.g., Google Books). This employed to retrieve and parse data from YouTube.com (Thelwall, 2012).
  217. 217. Social network analysis framework • (3) Communication patterns among international fans of Kpop across countries • Using webometric analyst, we collected data that related psy`s Gangnam style. It include video ID, published date, updated date, video title, video url, author name, dislike, likes viewcount, favorite count at al. • And most recent 1,000 comments posted to a GS video clips on Psy`s official Youtue acoount that uploaded on Psy's official YouTube account (“officialpsy”) was identified.
  218. 218. Social network analysis framework • A user-to-user network was constructed to reveal hidden relationships between commenters, i.e., nodes. Three networks of users were considered: a network of commentaries, a network of subscriptions, and subscriptions to a common network. Type Nodes refer to Ties occur when Commentary network Users commenting on the GS video. One user replies to a comment by another. Subscription network Same as above. One user subscribes to the channel/account of another. Subscriptions to a common network Same as above. Two users share common channel/account subscriptions on YouTube. Nodes and ties for each type of user network
  219. 219. Social network analysis framework • In terms of the geographical distribution of commenters, the U.S. had the largest number of commenters (46.93%, 214, N=456), followed by the U.K. (7.02%, 32), Canada (6.80%, 31), Korea (4.17%, 19), the Netherlands (2.85%, 13), Brazil (2.19%, 10), and Finland (2.19%, 10). • This reveals that Western users were influential in determining the flow of GS on YouTube. The sample was compared to demographics for all YouTube users in the U.S. According to Quantcast.com,
  220. 220. Results • This structural difference between the NC and the NSCN can be explained in part by the nature of YouTube. • In the Web 2.0 social media era, participants in internet forums are more synchronous by being more engaged in seeking information and selectively exposed to the congenial idea through receiving information highly personalized by their search and navigation patterns (Choi & Park, 2014). Types Commentary network Subscriptions to a common network Nodes 234 357 Ties 325 47,944 Density (Directed) 0.006 0.377 Density (Undirected) 0.010 0.377 Comparison of commentary networks and subscriptions to a common network in August
  221. 221. Figure1. Commentary network in August Gangnam Style Communication Networks on Youtube chain shape reflecting a circle
  222. 222. •. Figure 2. Subscriptions to a common network in August Gangnam Style Communication Networks on Youtube hub-and-spoke topology
  223. 223. • The structural pattern of the NC • Correlation analysis of common networks • These results indicate that frequent replies of commenters attracted some feedback from other commenters in the network because there was ongoing mutual recognition between repliers and those being replied to.t. Male users from the U.S. Outdegree IndegreeBinary Outde geeBin ary Indegree .546** .978** .506** Outdegree .487** .979** IndegreeBi nary .461**
  224. 224. In terms of the structural pattern of the NSCN, • According to the independent sample t-test, U.S. (N = 158) and non- U.S. (N = 180) commenters showed no difference in their channel co-subscription behaviors (undisclosed = 19) • Male commenters shared their subscription channels with others significantly more than female commenters. The average number of the shared subscription channels of male commenters was 58.40 (S.D. = 62.16), whereas that of female commenters, 43.00 (S.D. = 42.55).
  225. 225. Discussion & Implication • Asian popular music has grown rapidly, particularly in the U.S. and European countries, but such international diversity is not well reflected in the central channel for cultural discussions on music. The results have important implications for open digital settings, providing music firms with insights specifically focused on users' approaches (with mixed motives) to information dissemination. • Perhaps more importantly, the results have important practical implications for the music industry.
  226. 226. An analysis of Twitter communication on Organic products in Mexico and Korea using webometrics method. G.CD. Xanat V. Meza Advisor: Prof. Han Woo Park
  227. 227. Objectives • The present study compares social media resources for organic products between Mexico and Korea in the Twitter sphere in a period of six months. • A social media resource is any comment within or URL linked from a SNS page containing information on the production, consumption and diffusion of organic products (The Internet Society, 2005). Introduction
  228. 228. Literature Review Cross cultural research and SNS. • This study will apply a framework by Marcus & Gould (2001), which is based on Hofstede’s theory. • Several researchers (Ess & Sudweeks 2005, Callahan 2006, W¨urtz 2006, Gevorgyan & Manucharova 2009, Snelders, Morel & Havermans, 2011) have applied it to website features analyzes and users’ interaction.
  229. 229. Method Webometrics. “The study of web-based content with primarily quantitative methods for social science research goals and using techniques that are not specific to one field of study.” (Thelwall, 2009, p.6). “Hidden” and “relational” patterns can be discovered by extracting a sizeable quantity of data from the social media sphere. Webometrics could be particularly effective in identifying interrelationships between businesses’ stakeholders (Kim and Nam, 2012) .
  230. 230. Method Semantic analysis. • Analyses semantic relationships between concepts (Sowa, 1987). • In the present study, the unit of analysis is keywords.
  231. 231. Method Data collection procedures. • Hashtags for “Organic”: • Organico (in spanish) • 유기농 (in korean) • The process: • Collection of data by country • Classification of data by region. • Analysis of networks. • Classification of network influencers. • Analysis of TLDS. • Analysis and classification of linked URLs • Semantic analysis. • Analysis of hashtags and keywords.
  232. 232. Results RQ1.What is the diffusion path of social media resources for organic products in Mexico and Korea through Twitter? COUNTRY MX KOR Vertices 2382 7791 Total Edges 4227 37864 Maximum Geodesic Distance (Diameter) 20 15 Average Geodesic Distance 5.75 4.23 Average Betweenness Centrality 5848.87 23139.08
  233. 233. Results RQ1.1.How are the networks changing through time?
  234. 234. Results RQ1.1.How are the networks changing through time?
  235. 235. Results RQ1.1.How are the networks changing through time? 0 2000 4000 6000 8000 10000 January February March April May June Edges KOR Edges MX 0 500 1000 1500 2000 2500 Vertices KOR Vertices MX 0 1 2 3 4 5 6 7 Average geodesic distance KOR Average geodesic distance MX 0 5 10 15 20 Maximum geodesic distance KOR Maximum geodesic distance MX 0 1000 2000 3000 4000 5000 6000 7000 Average betweenn ess centrality KOR
  236. 236. Results RQ1.1.How are the networks changing through time? Correlations for Mexico Vertices Edges Maximum Geodesic Distance Average Geodesic Distance Betweenness Centrality Date 0.116 .203 -.053 -.019 .146 Significance .415 .149 .707 .891 .303 Correlations for Korea Vertices Edges Maximum Geodesic Distance Average Geodesic Distance Betweenness Centrality Date .449** .453** .253 .252 .289* Significance .001 .001 .070 .071 .037 Pearson correlation N = 52
  237. 237. Results RQ1.2. Who are influential players in diffusing organic products on Twitter?
  238. 238. Results RQ1.2. Who are influential players in diffusing organic products on Twitter?
  239. 239. Results RQ1.2. Who are influential players in diffusing organic products on Twitter? Indegree Centrality value Type of user Location Outdegree Centrality value Type of user Location KEN_QUOTES 136 General public Mexico City ExpoOrganicos 14 Business Mexico City mx_df 55 Alternative media Mexico City homeroblas 13 Celebrity Undefined En_laDelValle 46 Business Mexico City laorganizacion 13 Business Oaxaca PublimetroMX 40 Mass media Mexico City ChiczaMexico 13 Business Undefined tonygalifayad 37 Celebrity Puebla HacklCondesa 10 Business Mexico City laorganizacion 36 Business Oaxaca Tianguis_ 19 Business Mexico City Mean 58 Mean 25 Standard Deviation 38.691 Standard Deviation 6.022 Betweenness Centrality value Type of users Location Eigenvector Centrality value Type of users Location KEN_QUOTES 212381.479 General public Mexico City KEN_QUOTES 0.020 General public Mexico City ChiczaMexico 111712.703 Business Undefined ExpoOrganicos 0.010 Business Mexico City mx_df 98234.672 Alternative media Mexico City homeroblas 0.008 Celebrity Undefined ExpoOrganicos 97670.240 Business Mexico City laorganizacion 0.0007 Business Oaxaca laorganizacion 86222.745 Business Oaxaca mx_df 0.0006 Alternative media Mexico City anditagar 316512.780 General public Undefined ChiczaMexico .00006 Business Undefined Mean 430754 Mean 0.0066 Standard Deviation 88351.077 Standard Deviation 0.0058
  240. 240. Results RQ1.2. Who are influential players in diffusing organic products on Twitter? ALTERNATIVE MEDIA 1 POLITICIAN 2 BUSINESS 6 CITIZEN 2 MASS MEDIA 1
  241. 241. Results RQ1.2. Who are influential players in diffusing organic products on Twitter? Indegree Centrality value Type of user Location Outdegree Centrality value Type of user Location cjtlj 963 Business Undefined cjtlj 200 Business Undefined StarbucksKorea 368 Business Seoul GrouponKorea 125 Business Seoul wikitree 288 Alternative media Undefined doolbob 104 Alternative media Undefined six2k 245 General public Seoul erounnet 84 Mass media Undefined amazingkiss1104 237 General public Undefined sunshine7892 80 Business Gyeonggi Mangosix_kr 221 Business elelohemh 74 Business Gyeonggi Mean 387 Mean 111 Standard Deviation 287.109 Standard Deviation 47.381 Betweenness Centrality values Type of users Location Eigenvector Centrality values Type of users Location cjtlj 9497927.968 Business Undefined cjtlj 0.015 Business Undefined StarbucksKorea 3418206.580 Business Seoul Mangosix_kr 0.006 Business Undefined amazingkiss1104 3385445.805 General public Undefined StarbucksKorea 0.005 Business Seoul wikitree 3336795.105 Alternative media Undefined mosfkorea 0.004 Government Sejong six2k 2954885.522 General public Seoul melvita_korea 0.004 Business Seoul Sunshine7892 2082136.391 Business Gyeonggi busanbank 0.004 Business Busan Mean 4112566 Mean 0.0063 Standard Deviation 2686173.906 Standard Deviation 0.0043

×