Big Data and International

Relations
Andrej Zwitter*

F
rom November to , , ten heavily armed members of
Lashkar-e-Taiba (LeT), a Kashmiri separatist group, attacked several pub-
lic sites in Mumbai, India, with automatic weapons and grenades, killing
people and wounding three hundred. This was one of the rst known instanc-
es of terrorists employing powerful search algorithms such as Twitters or the link
analysis used in Googles PageRank system, which allowed LeT members to access
information from massive data pools in real-time. During the attacks, an LeT
operations center based in Pakistan communicated with the terrorists via sattelite
and GSM phones to provide them with open-source intelligence. From the oper-
ations center, LeT members data mined the Internet and social media, tapping
into the power of Big Data to provide the attackers with an intelligence advantage
over Indian law enforcement agencies. The attackers were thereby kept up to date
on the status of the Indian governments response and even received personal
proles of the hostages they took in the Taj Mahal Palace hotel.
Ironically, counterterrorism agencies are relying on the same technological
advances to stop similar attacks from occurring. For example, the U.S. military
is developing drones that, empowered by the combination of facial recognition
software and vast databases of face-tagged photos, are able to identify individuals
even in crowdsleading to what experts call hyperpersonalized warfare.
Furthermore, using data-mining techniques, trend spotting, and sentiment analy-
sis, experts in the area of predictive policing and intelligence analysis are hoping
to distill the indicators and to identify the anomalies that would help predict
terrorist attacks, counteract organized crime groups, and, at the same time, save
resourcesand potentially livesby employing targeted interventions.

*This essay was produced as part of the cooperation between the Austrian Institute for International Affairs and
the Danube University Krems, Austria.

Ethics & International Affairs, , no. (), pp. .

Carnegie Council for Ethics in International Affairs
doi:./S

377
 Big Data increasingly affects politics in manifold ways. With the ascendance of
cyberspace as an important domain of daily activity, international politics has al-
ready experienced technology-driven change. Big Data unveils new dimensions to
these changes, which we as political scientists and observers of international affairs
are only now beginning to comprehend. It changes power distributions and there-
by some basic assumptions of international relations theory, and its analytics will
increasingly inform international relations research and policymaking. It has cre-
ated both new opportunities and threats in areas such as humanitarian aid, devel-
opment, and international peace and security. As hardware becomes better and
cheaper, and as open-source software and database search and analysis services
become more widely available, the power of Big Data is also increasingly at the
disposal of small enterprises and individuals. Its ascendance in all aspects of social
and political life has also sharpened important questions about global Internet
governance.

What Is Big Data?

Big Data refers to the enormous amounts of data that, using sophisticated analyt-
ics techniques, can be mined for information in order to reveal patterns and spot
trends and correlations. A key idea behind the concept is that the sheer volume of data
allows users to discover informationspecically, correlations and patternsthat
would not be available by looking at smaller samples. It also relates to the enhanced
ability to extract information from, and interpret, massive amounts of unstructured
data. Another key idea is that Big Data is updated in near real-time. The most
important features relevant to understanding Big Data are known as the three
Vs, which are:

Volume: Data today exists in massive amounts, which can be measured in

petabytes [ bytes], exabytes [ bytes], and zetabytes [ bytes].
Soon, it will even be measured in yottabytes [], one of which equals
trillion digital video disks. According to Rick Smolan and Jennifer
Erwitt, from the beginning of history until , ve billion gigabytes
were produced; in this amount is produced every ten seconds.
Velocity: The speed of data creation and its collection now approaches
real-time. This concerns not only questions of bandwidth (megabyte and
gigabyte upload and download capabilities) but also of implementing

378 Andrej Zwitter

 information-technology architecture solutions that can cope with data in
near real-time.
Variety: Data exists in structured and unstructured forms and in different
formats and units of analysis, including documents, emails, social
media messages, YouTube videos, pictures, audio, radio-frequency identi-
cation chips, satellite imagery, sky cartography, DNA sequencing,
phone-network call data, and cell phone GPS signals. Furthermore, it
can be categorized depending on its source: for example, there is self-
generated data, data collected (mostly in automatized ways) from the
web (known as data scraping), and data retrieved from other outside
sources.
Some authors add a number of other features that are potentially, but not neces-
sarily, present in Big Data:

Veracity: This fourth V encompasses all sorts of methodological questions

about the reliability and validity of data and its sources. For example, pos-
sible biases can be generated by the form of collection (do sentiment anal-
yses generated from a certain region based on Twitter feeds actually
express the sentiments of that region, or only those of Twitter users
from that region?), and the form of data preparation (for example, by re-
moving duplicates, completing partial entries, aggregating results, and so
on).
Value: Oracle, for example, considers Big Data to be of low value density;
that is, the data received and collected requires much processing before
value can be extracted from it.
Correlations: Algorithms that analyze Big Data emphasize correlations
over causation for the use of predictions and for social engineering, that
is, the manipulation of individuals and groups based on social
mechanisms.
Exhaustiveness: Related to Volume, Big Data becomes potentially
all-encompassing of a research population, unlike with statistics, which
commonly works with samples that only represent and approximate the
wholethat is, the sample size is moving toward a state of N = all.
Detailed/Organic Data: Unlike Exhaustiveness, which is about sample size,
organic data refers to the degree of granularity of the data within the
sample; increasingly, organic data allows for a more accurate digital

big data and international relations 379

 representation of physical reality in such ne detail that it approaches an
organic representation of realitymetaphorically, a : map of the world.
Flexibility: Big Data encompasses extensionality (new dimensions can be
added by adding new datasets) and scalability (the size of datasets can
be expanded rapidly).
Virality: This factor measures how quickly data is spread and shared across
networks of people (PP) to each unique node. The time and rate of the
spread of information are the determinant factors. Viral diffusion differs
from broadcasting in that the former, acting through a network structure,
allows its individual nodes to contribute to social cascades, resulting in
viral stories and posts.
In sum, Big Data promises an information advantage, be it in business intelli-
gence, state intelligence, or any other form of data gathering and analysis. At
least in theory, it offers the ability to become omniscient, if not omnipotenta
promise that is tempting to any business, government, or criminal network.

New Actors and New Power Distributions

New Actors
Mainstream international relations theories of the past decades have emphasized
structural explanations that privilege states and intergovernmental organizations
as the primary agents of international phenomena. In these models, corporate ac-
tors as well as individuals or substate groups play a much smaller role or no role at
all. These theories, however, have been challenged by the emergence of cyberspace
and its capacity to empower individual action. Notably, small substate interest
groups, such as those that spearheaded the Arab Spring and Occupy movements,
gained traction because of their members ability to bridge the cyber-physical gap.
Big Data is adding new dimensions to these transformations as it changes the
power distribution among various actors. It is crucial to understand that Big
Data is now commonly viewed as the new oila raw material that, if rened prop-
erly, has immense value. In response, a host of new actors have arisen specializing
in the collection of information from the Internet, the buying and selling of con-
sumer data, analytics and visualization, and so on.
As individuals digital footprints expandthrough their use of credit cards,
bonus/loyalty cards, cell phones, and the Internet of Things (that is, the network
of everyday objects, from smart watches to rdges to cars, embedded with low-cost

380 Andrej Zwitter

sensor technology that enables them to exchange data)the power of these new
actors only increases. Big Data collectors such as social media providers, search
engines, banks, and marketing and IT companies determine which data is collect-
ed, what is stored and how, and for how long. When it comes to the quality and
veracity of this data, buyersincluding governmentsare at the mercy of these
very same data-mining companies and data brokers.
Another new type of stakeholder are Big Data utilizers, such as administrative
and intelligence agencies, and companies that seek to improve their services.
Utilizers reassemble different datasets and databases, (re)dening the purposes
for data. In some cases, we nd collectors and utilizers combined into one
actor, such as Google, Microsoft, Cisco, and many smaller companies. In other
cases, companies buy data sets from intermediary data brokers (such as Acxiom
and Intelius) for their own analysis needs. All these actors have a great inuence
over the production and dissemination of knowledge and innovation.
Governments are increasingly outsourcing their data collection, and sometimes
even their data analysis tasks, to private actors. For example, the telecom company
AT&T has reportedly been paid $ million for storing and providing data to U.S.
intelligence agencies.
Since storage and processing power have become increasingly affordable, more
and more small IT companies with a focus on different areas of the Big Data value
chain (collection, data preparation, analysis, visualization) are emerging. Certain
Big Data pools are readily accessible to anyone with a laptop or mobile device.
Even if the possibility to repurpose Big Data sets remains with the Big Data uti-
lizers, individuals can now access Big Data with varying degrees of sophistication
without needing to possess the most powerful and expensive hardware. For exam-
ple, with its BigQuery service, Google provides its own processing power to its
customers. However, as long as most of the processing power, data gathering ca-
pabilities, and know-how in Big Data analysis remains in the hands of big com-
panies, they will retain a distinctive competitive advantage on knowledge
production and innovation vis--vis both individuals and states.

Individual Agency
It is well known that the Internet has revolutionized how we communicate and
has empowered individuals: with the rise of online substate groups, individual
agency seems to be ourishing. Big Data, on the other hand, pushes individual
agency into the background. A recent case helps to illustrate this trend.

big data and international relations 381

 In May the website PatientsLikeMe.comwhere people can share infor-
mation and concerns about their diseases, exchange treatment protocols, and dis-
cuss medication recommendationsdiscovered a major intrusion. The Nielsen
company, through its subsidiary BuzzMetrics, had scraped all data from the
forum, including otherwise secret medical information. Nielsen subsequently ad-
mitted that it was mining data from million blogs, eight thousand message
boards, and several social media sites in order to sell it to advertisers, marketers,
and pharmaceutical companies. Indeed, the ordinary user of Facebook, Gmail,
Yahoo, and Twitteror even just a cell phone or credit cardhas little to no
choice about what data will be collected on her or him. National and regional
data protection mechanisms seem to be of limited help. One is reminded of
Lawrence Lessigs comment that, just as human conduct in the physical domain
is regulated by laws between the citizen and the state, human conduct in the
cyber domain is regulated by computer codes most often developed by corporate
agencies.

Correlation over Causation

One specic feature of Big Data is already progressively informing international
policy and agenda settingits privileging of correlations over causations. A fre-
quent argument is that the fact that we are now dealing with nearly complete
data sets (N = all) leads to the redundancy of theory. In other words, some have ar-
gued that correlations would replace causal models for good. Such Big Data cor-
relations already show their usefulness for policymaking. For example, the UN
Global Pulse projecta Big Data-based initiative that tracks real-time developments
regarding human wellbeing and vulnerabilitiesfound a strong correlation between
Twitter conversations about the price of food and food price ination. The Global
Pulse Lab Jakarta, for example, mined Indonesian price-related Twitter data and ap-
plied a classication algorithm in order to conduct a sentiment analysis and corre-
late it with ofcial food ination statistics. If things develop in accordance with the
hopes of the UN Global Pulse, in the future real-time access to macrodata analytics
will help inform policymaking, allowing ad hoc adjustments to be made with a solid
evidence base. Researchers are also using trend spotting and sentiment analysis to
detect emergent political conicts through web mining.
One can expect that this sort of evidence-based policymaking will become more
important as Big Data analysis approaches real-time. Still, while we can anticipate
interesting empirical ndings from Big Data-enhanced research, it is unlikely that

382 Andrej Zwitter

it will replace theory and model building in the study of international relations.
Big Data correlations sometimes do not meet the standards of scientic research,
as the failure of Google Flu Trends shows. Moreover, knowing that certain var-
iables correlate does not say anything about the nature of a possible causal rela-
tionship, or whether one is even present.

Opportunities of the Big Data Age

Humanitarian and development aid agencies are dependent on reliable data for
operational planning, logistics, and monitoring of their projects. In order to facil-
itate these processes, several platforms have been developed to provide humanitar-
ian aid workers with open access to information, with many more on the way.
This sharing of data can help facilitate aid delivery by NGOs that otherwise do
not have the capability to gather intelligence on the ground themselves. For exam-
ple, when a disaster strikes, humanitarian operational planning based on a rapid
needs assessment should usually be ready within seventy-two hours. Using Big
Data, a number of organizations have tried to help narrow this information gap.
Consider the Kenyan open-source software company Ushahidi. Ushahidi offers
users the ability to upload real-time information about crises from a GPS-enabled
cell phone onto a map that is freely accessible. It was used extensively to support
humanitarian operations after the Haiti earthquake, and it improved the
election monitoring in Kenya in (known as the Uchaguzi project), among
many other examples. Since most crisis-mapping tools merely consist of a web-
site accessible with any Internet-enabled device, Ushahidi can provide aid workers
with real-time intelligence in the palm of their hands. Similarly, other crisis map-
pers, such as the Standby Task Force and Sahanna, use overlay technology (often
provided by volunteers) to project crowd-sourced information onto maps to in-
form humanitarian and development aid workers.
International nongovernmental organizations are also experimenting with this
powerful new technology. For instance, the Humanitarian Data Exchange (HDX),
a project launched by the United Nations Ofce for the Coordination of
Humanitarian Affairs, has already assembled more than , data sets. One ag-
gregates the number of Ebola cases and infected aid workers; another monitors
water sources and opinions about water quality in Kenya; and another keeps
track of total uniformed personnel of each contributing member-country by
month, type (troop, police, or expert/observer), and mission.

big data and international relations 383

 With regard to national and international security, Big Data already plays an
important role. For example, percent of the police forces surveyed across the
United States are actively using social media, including for investigative and intel-
ligence purposes. One of the recent advances of Big Data in national security
and public safety can be seen in the area of predictive policing, which is the use
of data to forecast where and at what time crimes are more likely to occur.
This analysis can then be used to keep certain areas under surveillance and to pre-
ventively dispatch police units.
In October , the UN Secretary-General released a report detailing how the
use of technology has become an intrinsic part of decision-making, information
gathering, and planning in support of the United Nations core mandates in the
areas of peace and security, development, human rights, and international law.
In the words of the report: The increasing availability of free and open-source
software and open data reinforces the need for the United Nations to collaborate
in the movement of big data, joined with humanitarian and social networking.
For that purpose, the Secretary-General laid out a strategy to provide a common
vision for the delivery of information and communications technology throughout
the United Nations system, focusing on improving the organizations internal in-
formation, communication, and technology capabilities and increasingly investing
in Big Data innovations.
Researchers are already envisioning the role that Big Data can play in the future
for UN peacekeeping and peacebuilding. John Karlsrud, for example, argues that
the ongoing cyberization of conict prevention, humanitarian action, and devel-
opment, and the possibility of tracking population ows, will lead to a new gen-
eration of peacekeeping (peacekeeping .) and peacebuilding that embraces
these new advancements in real-time awareness, feedback, and early-warning
systems.
Satellite imagery is increasingly becoming another source of Big Data analytics.
Location data correlated with surveys, photos, and maps can help in uncovering
war crimes. For example, in the Satellite Sentinel Programme of the
Harvard Humanitarian Initiative reported nding eight mass graves in and
around Kadugli, Sudan, with the help of analyzed DigitalGlobe imagery corrobo-
rated with details retrieved from UN reports and eyewitness accounts.
Big Data may also be able to predict and forecast conicts and social instabil-
ities. For example, the U.S. Defense Departments Information Volume and
Velocity program aims to harness strategic intelligence from Big Data by using

384 Andrej Zwitter

pattern recognition to detect social instabilities in populations. And, as mentioned
above, the United Nations has initiated the Global Pulse project to track real-time
developments regarding human wellbeing and vulnerabilities. These kinds of pro-
grams have led some commentators to argue that Big Data might be able to help
prevent future conicts.

Emerging Risks and Challenges

Big Data, however, also possesses huge potential for misuse. Humanitarian and
development aid workers are at a constant risk of attack by ideologically motivated
groups, but also by criminals seeking prot. Traveling with money, very little
security (many aid organizations consider military escorts a violation of human-
itarian principles), and expensive equipment, they are easy targets. Ushahidi and
other crowd-sourced crisis mapping software provide more detailed and better
structured human and signals intelligence (HUMINT and SIGINT) than criminals
and armed groups could have ever hoped for just a few years ago. Crisis mappers
learned this lesson when responding to the oods and food crisis in Pakistan.
Pakistan-based Taliban forces threatened to attack all foreign aid workers, such as
those with the World Food Programme or Mdecins Sans Frontires, and the team
operating the Pakreport.org crisis map had to rapidly change its approach to avoid
allowing open-access data to be used to target humanitarians.
Crisis mappers have become increasingly aware of the potential misuse of this
technology. Consequently, when in the United Nations requested a group of
researchers to develop a platform to map the Libya crisis, the mappers created a
password-protected version of the site. Big Data also reduces the individuals
ability to foresee the consequences of his or her actions. This can have political
consequences, as was illustrated by the case of the Russian soldier who posted
geotagged photos online, indicating that, despite President Vladimir Putins
assurances, Russian soldiers were in Ukraine.
The way Big Data is used to generate insights about individuals and groups cre-
ates new types of challenges. Already, Big Data analysis on the basis of group
behavior, preferences, likes/dislikes, and so on informs marketers in near real-
time. At the same time, websites and web services (including Google Search)
are increasingly tailored to individuals. Information gleaned from a persons
prior online behavior is now employed to, say, feature certain goods available
for purchase. It even determines which news items are prioritized in a search

big data and international relations 385

engine or social feed. The term for this phenomenon is ambient intelligence, and
it is quietly invading all areas of technology. While personalized information can
be a service, it also exposes individuals vulnerabilities that can be exploited for the
purpose of social engineering and manipulation.
The rise of Big Data also presents other, fundamental, epistemic challenges. The
moral and legal frameworks pioneered during the Enlightenment and codied in
the postwar era are, like human rights, inherently designed around the individual
actor and her specic individualistic interests, such as privacy. These norms, how-
ever, are not adequate when it comes to the privacy of groups. Many also raise
concerns that this focus on groups and group behavior might lead to racial pro-
ling, specically in the context of predictive policing.
Finally, the general public does not understand the Big Data phenomenon well
enough to be particularly concerned by it, or to demand of their government spe-
cic protections from its implications. Of course, whistleblowers and investigative
journalists occasionally offer glimpses into the reality of the cyber domain. But
this remains insufcient to create a general public awareness of the actual digital
footprint our Internet presence and cell-phone use creates. Even if the right to be
forgotten makes sense in certain cases, applying it as an individual to one com-
pany at a time still leaves unaffected the many other data collectors and brokers
that are collecting the same data and of whose activities we are not aware.

Governance of Big Data

Big Data is increasingly important in all domains of social and political life, which
suggests the need for governance to curb potential abuses. In the cyber domain,
actors nd increasing space to steer free of limitations imposed by national legis-
lationsa case in point is the recently dismantled website Silk Road, part of the
so-called dark web, which trafcked extensively in weapons and drugs. Big
Data accentuates this trend of ungoverned space, as practices such as data collec-
tion and data mining are inherently global in reach. It comes as no surprise that
companies and data warehouses use a variety of strategies to circumvent national
and regional legislation that might limit these practices.
Take the Austrian initiative Europe-v-Facebook.org, which details Facebooks
obstructive strategies in delaying the case of , European citizens trying to
bring their data protection case to the European Court of Justice and other nation-
al courts. For its part, Google has tried to undermine EU data protection

386 Andrej Zwitter

measures, such as those related to the right to be forgotten. Paul Nemitz, a director
in the European Commissions justice department, has even argued that Google
used meetings of its advisory council in Europe as a passive-aggressive PR strategy
to delegitimize the EUs data protection rules and jurisprudence. National legis-
lations are not much better equipped to tackle these challenges: a look at the
Global Data Protection Handbook reveals that few states have adequate data pro-
tection laws or any such laws at all.
Is effective international regulation, for example by an international treaty, pos-
sible? It takes just one outlier state that does not sign such a treaty on Big Data
governance to jeopardize the whole endeavor. The question might be whether self-
regulation, by way of a voluntary code of ethics, is a viable alternative. Such codes
of conduct employ a soft form of governance that can sometimes be just as effec-
tive as laws and regulations. Not abiding by such codes can have consequences,
such as public criticism, a loss of customers, andif lawmakers and funders
adopt these codes as benchmarksthe possible loss of funding and market oppor-
tunities. Of course, the creation of such a code would require the initiative of glob-
al players, civil society organizations, and academia; and these principles and
decision-making heuristics would need to be open enough so as not to hamper
innovation while still safeguarding against the abovementioned threats.

Conclusion
In their book Cyberspace and International Relations, Jan-Frederik Kremer and
Benedikt Mller argue that the global cyberspace substrate has undermined the
older distinctions between international and domestic, between peace and war, be-
tween state and non-state actors, and between technology, politics, and econom-
ics. I would go even further: global cyberspace, and its intensication through
the ubiquitous nature of Big Data and the Internet of Things, challenges us to re-
think fundamental notions of international relations and powernotions that we
have taken for granted for decades.
There are many challenges ahead. Governance efforts should strive to reduce
sensitive data and exploitable information from becoming open data.
Furthermore, despite ongoing efforts by the European Union and other state ac-
tors to update data protection regulationsan effort that is lagging behind relent-
less technological innovationthe governance of nonstate actors has largely
escaped national and international legislation.

big data and international relations 387

 When it comes to eventually developing legal frameworks, the nature of Big
Data and the modus operandi of actors in the cyber domain increasingly show
that the current principles guiding national and international legislation are insuf-
cient. Groups will increasingly require their own legal protection mechanisms. In
order to reduce the vulnerabilities that Big Data has imposed on society, knowl-
edgeable stakeholders need to raise the awareness of people about the pitfalls
that Big Data and their own digital footprint bring. Big Data certainly holds the
promise for improving global wellbeing, and perhaps even for preventing conicts.
Nevertheless, it can also be the source of much evil. The only way to keep the mis-
use of Big Data in check is to create a global awareness and to use the tools that
Big Data itself offerssocial media and global connectednessto enable civil so-
ciety to become the public watchdog of the Big Data age.

NOTES

Marc Goodman, Future Crimes: Everything Is Connected, Everyone Is Vulnerable, and What We Can Do
About It, ebook (New York: Knopf Doubleday Publishing Group, ), ch. .

Charles J. Dunlap Jr., The Hyper-Personalization of War: Cyber, Big Data, and the Changing Face of
Conict, Georgetown Journal of International Affairs , International Engagement on Cyber IV
(), pp. .

Rick Smolan and Jennifer Erwitt, The Human Face of Big Data (Sausalito, Calif.: Against All Odds
Productions, ); Rob Kitchin, Big Data and Human Geography: Opportunities, Challenges and
Risks, Dialogues in Human Geography , no. (), pp. ; Doug Laney, D Data
Management: Controlling Data Volume, Velocity, and Variety (Meta Group, February , ),
blogs.gartner.com/doug-laney/les///ad-D-Data-Management-Controlling-Data-Volume-
Velocity-and-Variety.pdf; and Bill Vorhies, How Many Vs in Big DataThe Characteristics That
Dene Big Data, Business Foundation Series #, October , , data-magnum.com/how-many-
vs-in-big-data-the-characteristics-that-dene-big-data/.

Ibid.

Small chips attached to objects that contain electronically stored and wirelessly transferred information,
e.g., for tracking and identifying parcels (functionally similar to QR codes).

Richard Winter, Big Data: Business Opportunities, Requirements, and Oracles Approach, Executive
Report, Winter Corporation, December , p. , www.oracle.com/us/corporate/analystreports/infras
tructure/winter-big-data-.pdf.

Andrej Zwitter, Big Data Ethics, Big Data & Society , no. (), p. ; Viktor Mayer-Schnberger
and Kenneth Cukier, Big Data: A Revolution That Will Transform How We Live, Work, and Think
(Boston: Houghton Mifin Harcourt, ), p. ff.

Mayer-Schnberger and Cukier, Big Data, pp. .

Zwitter, Big Data Ethics, p. .

Rob Kitchin, Big Data, New Epistemologies and Paradigm Shifts, Big Data & Society , no. (),
p. .

Ray Wang, Mondays Musings: Beyond The Three Vs of Big DataViscosity and Virality, Forbes,
February , .

Lilian Weng, Filippo Menczer, and Yong-Yeol Ahn, Virality Prediction and Community Structure in
Social Networks, Scientic Reports , Article number: (published online August , ).

Jacob Silverman, Time to Regulate Data Brokers, Al Jazeera America, Opinion section, January ,
, america.aljazeera.com/opinions///time-to-regulatedatabrokers.html.

Goodman, Future Crimes, ch. .

Lawrence Lessig, Code and Other Laws of Cyberspace (New York: Basic Books, ).

Chris Anderson, The End of Theory: The Data Deluge Makes the Scientic Method Obsolete,
WIRED, June , , archive.wired.com/science/discoveries/magazine/-/pb_theory/.

388 Andrej Zwitter


F. Johansson et al., Detecting Emergent Conicts through Web Mining and Visualization, Intelligence
and Security Informatics Conference (EISIC), European, pp. .

David Lazer et al., The Parable of Google Flu: Traps in Big Data Analysis, Science , no.
(), pp. .

Ushahidi, www.ushahidi.com/, accessed April , ; and Kenyan Elections , Community

WikiUshahidi, wiki.ushahidi.com/display/WIKI/Uchaguzi+-+Kenyan+Elections+, accessed April
, .

Center for Social Media, Social Media Survey Results, Annual Survey (Alexandria, Va.:
International Association of Chiefs of Police, Fall ), www.iacpsocialmedia.org/Resources/Publica-
tions/SurveyResults.aspx.

UN Secretary-General, Information and Communications Technology in the United Nations, Report

of the Secretary-General to the General Assembly, October , , UN document A//, para. .

John Karlsrud, Peacekeeping .: Harnessing the Potential of Big Data, Social Media, and Cyber
Technologies, in Jan-Frederik Kremer and Benedikt Mller, eds., Cyberspace and International
Relations: Theory, Prospects and Challenges, edition (Heidelberg, Ger.: Springer, ), pp. .

Satellite Sentinel Project, Evidence of Burial of Human Remains in Kadugli, South Kordofan, Special
Report, Harvard Humanitarian Initiative, August , , hhi.harvard.edu/publications/special-
report-evidence-burial-human-remains-kadugli-south-kordofan.

Sheldon Himelfarb, Can Big Data Stop Wars Before They Happen? Foreign Policy, April , ,
foreignpolicy.com////can-big-data-stop-wars-before-they-happen/.

Rasool Dawar (Associated Press), Taliban Threatens Foreign Aid Workers, Washingtion Times
online, August , , www.washingtontimes.com/news//aug//taliban-threatens-foreign-
aid-workers/.

Marc Parry, Academics Join Relief Efforts Around the World as Crisis Mappers, Chronicle of Higher
Education, March , , chronicle.com/article/Academics-Join-Relief-Efforts//.

Mireille Hildebrandt, Smart Technologies and the End(s) of Law: Novel Entanglements of Law and
Technology (Cheltenham, U.K.: Edward Elgar Pub, ), p. .

John J. Reilly Center of the University of Notre Dame, Predictive Policing, reilly.nd.edu/
outreach/emerging-ethical-dilemmas-and-policy-issues-in-science-and-technology-/predictive-policing/.

I.e., a website that cannot be indexed by search engines and to which one only has access through
identity-cloaking protocols such as Tor.

Europe-v-Facebook.org, www.europe-v-facebook.org, accessed August , .

Reuters, Google Accused of Passive-Aggressiveness over EU Right to Be Forgotten, Telegraph,

Technology section, November , , www.telegraph.co.uk/technology/google//Google-
accused-of-passive-aggressiveness-over-EU-right-to-be-forgotten.html.

Global Data Protection Handbook, dlapiperdataprotection.com/#handbook/world-map-section/c_

AR/c_EG, accessed April , .

Jan-Frederik Kremer and Benedikt Mller, eds., Cyberspace and International Relations: Theory,
Prospects and Challenges, edition (Heidelberg, Ger.: Springer, ), p. vii.

big data and international relations 389