Category Archives: Crowdsourcing

How to Create Resilience Through Big Data

Revised! I have edited this article several dozen times since posting the initial draft. I have also made a number of substantial changes to the flow of the article after discovering new connections, synergies and insights. In addition, I  have greatly benefited from reader feedback as well as the very rich conversa-tions that took place during the PopTech & Rockefeller workshop—a warm thank you to all participants for their important questions and feedback!

Introduction

I’ve been invited by PopTech and the Rockefeller Foundation to give the opening remarks at an upcoming event on interdisciplinary dimensions of resilience, which is  being hosted at Georgetown University. This event is connected to their new program focus on “Creating Resilience Through Big Data.” I’m absolutely de-lighted to be involved and am very much looking forward to the conversations. The purpose of this blog post is to summarize the presentation I intend to give and to solicit feedback from readers. So please feel free to use the comments section below to share your thoughts. My focus is primarily on disaster resilience. Why? Because understanding how to bolster resilience to extreme events will provide insights on how to also manage less extreme events, while the converse may not be true.

Big Data Resilience

terminology

One of the guiding questions for the meeting is this: “How do you understand resilience conceptually at present?” First, discourse matters.  The term resilience is important because it focuses not on us, the development and disaster response community, but rather on local at-risk communities. While “vulnerability” and “fragility” were used in past discourse, these terms focus on the negative and seem to invoke the need for external protection, overlooking the fact that many local coping mechanisms do exist. From the perspective of this top-down approach, international organizations are the rescuers and aid does not arrive until these institutions mobilize.

In contrast, the term resilience suggests radical self-sufficiency, and self-sufficiency implies a degree of autonomy; self-dependence rather than depen-dence on an external entity that may or may not arrive, that may or may not be effective, and that may or may not stay the course. The term “antifragile” just recently introduced by Nassim Taleb also appeals to me. Antifragile sys-tems thrive on disruption. But lets stick with the term resilience as anti-fragility will be the subject of a future blog post, i.e., I first need to finish reading Nassim’s book! I personally subscribe to the following definition of resilience: the capacity for self-organization; and shall expand on this shortly.

(See the Epilogue at the end of this blog post on political versus technical defini-tions of resilience and the role of the so-called “expert”. And keep in mind that poverty, cancer, terrorism etc., are also resilient systems. Hint: we have much to learn from pernicious resilience and the organizational & collective action models that render those systems so resilient. In their book on resilience, Andrew Zolli and Ann Marie Healy note the strong similarities between Al-Qaeda & tuber-culosis, one of which are the two systems’ ability to regulate their metabolism).

Hazards vs Disasters

In the meantime, I first began to study the notion of resilience from the context of complex systems and in particular the field of ecology, which defines resilience as “the capacity of an ecosystem to respond to a perturbation or disturbance by resisting damage and recovering quickly.” Now lets unpack this notion of perturbation. There is a subtle but fundamental difference between disasters (processes) and hazards (events); a distinction that Jean-Jacques Rousseau first articulated in 1755 when Portugal was shaken by an earthquake. In a letter to Voltaire one year later, Rousseau notes that, “nature had not built [process] the houses which collapsed and suggested that Lisbon’s high population density [process] contributed to the toll” (1). In other words, natural events are hazards and exogenous while disas-ters are the result of endogenous social processes. As Rousseau added in his note to Voltaire, “an earthquake occurring in wilderness would not be important to society” (2). That is, a hazard need not turn to disaster since the latter is strictly a product or calculus of social processes (structural violence).

And so, while disasters were traditionally perceived as “sudden and short lived events, there is now a tendency to look upon disasters in African countries in particular, as continuous processes of gradual deterioration and growing vulnerability,” which has important “implications on the way the response to disasters ought to be made” (3). (Strictly speaking, the technical difference between events and processes is one of scale, both temporal and spatial, but that need not distract us here). This shift towards disasters as processes is particularly profound for the creation of resilience, not least through Big Data. To under-stand why requires a basic introduction to complex systems.

complex systems

All complex systems tend to veer towards critical change. This is explained by the process of Self-Organized Criticality (SEO). Over time, non-equilibrium systems with extended degrees of freedom and a high level of nonlinearity become in-creasingly vulnerable to collapse. Social, economic and political systems certainly qualify as complex systems. As my “alma mater” the Santa Fe Institute (SFI) notes, “The archetype of a self-organized critical system is a sand pile. Sand is slowly dropped onto a surface, forming a pile. As the pile grows, avalanches occur which carry sand from the top to the bottom of the pile” (4). That is, the sand pile becomes increasingly unstable over time.

Consider an hourglass or sand clock as an illustration of self-organized criticality. Grains of sand sifting through the narrowest point of the hourglass represent individual events or natural hazards. Over time a sand pile starts to form. How this process unfolds depends on how society chooses to manage risk. A laisser-faire attitude will result in a steeper pile. And grain of sand falling on an in-creasingly steeper pile will eventually trigger an avalanche. Disaster ensues.

Why does the avalanche occur? One might ascribe the cause of the avalanche to that one grain of sand, i.e., a single event. On the other hand, a complex systems approach to resilience would associate the avalanche with the pile’s increasing slope, a historical process which renders the structure increasingly vulnerable to falling grains. From this perspective, “all disasters are slow onset when realisti-cally and locally related to conditions of susceptibility”. A hazard event might be rapid-onset, but the disaster, requiring much more than a hazard, is a long-term process, not a one-off event. The resilience of a given system is therefore not simply dependent on the outcome of future events. Resilience is the complex product of past social, political, economic and even cultural processes.

dealing with avalanches

Scholars like Thomas Homer-Dixon argue that we are becoming increasingly prone to domino effects or cascading changes across systems, thus increasing the likelihood of total synchronous failure. “A long view of human history reveals not regular change but spasmodic, catastrophic disruptions followed by long periods of reinvention and development.” We must therefore “reduce as much as we can the force of the underlying tectonic stresses in order to lower the risk of synchro-nous failure—that is, of catastrophic collapse that cascades across boundaries between technological, social and ecological systems” (5).

Unlike the clock’s lifeless grains of sand, human beings can adapt and maximize their resilience to exogenous shocks through disaster preparedness, mitigation and adaptation—which all require political will. As a colleague of mine recently noted, “I wish it were widely spread amongst society  how important being a grain of sand can be.” Individuals can “flatten” the structure of the sand pile into a less hierarchical but more resilience system, thereby distributing and diffusing the risk and size of an avalanche. Call it distributed adaptation.

operationalizing resilience

As already, the field of ecology defines  resilience as “the capacity of an ecosystem to respond to a perturbation or disturbance by resisting damage and recovering quickly.” Using this understanding of resilience, there are at least 2 ways create more resilient “social ecosystems”:

  1. Resist damage by absorbing and dampening the perturbation.
  2. Recover quickly by bouncing back or rather forward.

Resisting Damage

So how does a society resist damage from a disaster? As hinted earlier, there is no such thing as a “natural” disaster. There are natural hazards and there are social systems. If social systems are not sufficiently resilient to absorb the impact of a natural hazard such as an earthquake, then disaster unfolds. In other words, hazards are exogenous while disasters are the result of endogenous political, economic, social and cultural processes. Indeed, “it is generally accepted among environmental geographers that there is no such thing as a natural disaster. In every phase and aspect of a disaster—causes, vulnerability, preparedness, results and response, and reconstruction—the contours of disaster and the difference between who lives and dies is to a greater or lesser extent a social calculus” (6).

So how do we apply this understanding of disasters and build more resilient communities? Focusing on people-centered early warning systems is one way to do this. In 2006, the UN’s International Strategy for Disaster Reduction (ISDR) recognized that top-down early warning systems for disaster response were increasingly ineffective. They thus called for a more bottom-up approach in the form of people-centered early warning systems. The UN ISDR’s Global Survey of Early Warning Systems (PDF), defines the purpose of people-centered early warning systems as follows:

“… to empower individuals and communities threatened by hazards to act in sufficient time and in an appropriate manner so as to reduce the possibility of personal injury, loss of life, damage to property and the environment, and loss of livelihoods.”

Information plays a central role here. Acting in sufficient time requires having timely information about (1) the hazard/s, (2) our resilience and (3) how to respond. This is where information and communication technologies (ICTs), social media and Big Data play an important role. Take the latter, for example. One reason for the considerable interest in Big Data is prediction and anomaly detection. Weather and climatic sensors provide meteorologists with the copious amounts of data necessary for the timely prediction of weather patterns and  early detection of atmospheric hazards. In other words, Big Data Analytics can be used to anticipate the falling grains of sand.

Now, predictions are often not correct. But the analysis of Big Data can also help us characterize the sand pile itself, i.e., our resilience, along with the associated trends towards self-organized criticality. Recall that complex systems tend towards instability over time (think of the hourglass above). Thanks to ICTs, social media and Big Data, we now have the opportunity to better characterize in real-time the social, economic and political processes driving our sand pile. Now, this doesn’t mean that we have a perfect picture of the road to collapse; simply that our picture is clearer than ever before in human history. In other words, we can better measure our own resilience. Think of it as the Quantified Self move-ment applied to an entirely different scale, that of societies and cities. The point is that Big Data can provide us with more real-time feedback loops than ever before. And as scholars of complex systems know, feedback loops are critical for adaptation and change. Thanks to social media, these loops also include peer-to-peer feedback loops.

An example of monitoring resilience in real-time (and potentially anticipating future changes in resilience) is the UN Global Pulse’s project on food security in Indonesia. They partnered with Crimson Hexagon to forecast food prices in Indonesia by analyzing tweets referring to the price of rice. They found an inter-esting relationship between said tweets and government statistics on food price inflation. Some have described the rise of social media as a new nervous system for the planet, capturing the pulse of our social systems. My colleagues and I at QCRI are therefore in the process of appling this approach to the study of the Arabic Twittersphere. Incidentally, this is yet another critical reason why Open Data is so important (check out the work of OpenDRI, Open Data for Resilience Initiative. See also this post on Demo-cratizing ICT for Development with DIY Innovation and Open Data). More on open data and data philanthropy in the conclusion.

Finally, new technologies can also provide guidance on how to respond. Think of Foursquare but applied to disaster response. Instead of “Break Glass in Case of Emergency,” how about “Check-In in Case of Emergency”? Numerous smart-phone apps such as Waze already provide this kind of at-a-glance, real-time situational awareness. It is only a matter of time until humanitarian organiza-tions develop disaster response apps that will enable disaster-affected commu-nities to check-in for real time guidance on what to do given their current location and level of resilience. Several disaster preparedness apps already exist. Social computing and Big Data Analytics can power these apps in real-time.

Quick Recovery

As already noted, there are at least two ways create more resilient “social eco-systems”. We just discussed the first: resisting damage by absorbing and dam-pening the perturbation.  The second way to grow more resilient societies is by enabling them to rapidly recover following a disaster.

As Manyena writes, “increasing attention is now paid to the capacity of disaster-affected communities to ‘bounce back’ or to recover with little or no external assistance following a disaster.” So what factors accelerate recovery in eco-systems in general? In ecological terms, how quickly the damaged part of an ecosystem can repair itself depends on how many feedback loops it has to the non- (or less-) damaged parts of the ecosystem(s). These feedback loops are what enable adaptation and recovery. In social ecosystems, these feedback loops can be comprised of information in addition to the transfer of tangible resources.  As some scholars have argued, a disaster is first of all “a crisis in communicating within a community—that is, a difficulty for someone to get informed and to inform other people” (7).

Improving ways for local communities to communicate internally and externally is thus an important part of building more resilient societies. Indeed, as Homer-Dixon notes, “the part of the system that has been damaged recovers by drawing resources and information from undamaged parts.” Identifying needs following a disaster and matching them to available resources is an important part of the process. Indeed, accelerating the rate of (1) identification; (2) matching and, (3) allocation, are important ways to speed up overall recovery.

This explains why ICTs, social media and Big Data are central to growing more resilient societies. They can accelerate impact evaluations and needs assessments at the local level. Population displacement following disasters poses a serious public health risk. So rapidly identifying these risks can help affected populations recover more quickly. Take the work carried out by my colleagues at Flowminder, for example. They  empirically demonstrated that mobile phone data (Big Data!) can be used to predict population displacement after major disasters. Take also this study which analyzed call dynamics to demonstrate that telecommunications data could be used to rapidly assess the impact of earthquakes. A related study showed similar results when analyzing SMS’s and building damage Haiti after the 2010 earthquake.

haiti_overview_570

Resilience as Self-Organization and Emergence

Connection technologies such as mobile phones allow individual “grains of sand” in our societal “sand pile” to make necessary connections and decisions to self-organize and rapidly recover from disasters. With appropriate incentives, pre-paredness measures and policies, these local decisions can render a complex system more resilient. At the core here is behavior change and thus the importance of understanding behavior change models. Recall  also Thomas Schelling’s observation that micro-motives can lead to macro-behavior. To be sure, as Thomas Homer-Dixon rightly notes, “Resilience is an emergent property of a system—it’s not a result of any one of the system’s parts but of the synergy between all of its parts.  So as a rough and ready rule, boosting the ability of each part to take care of itself in a crisis boosts overall resilience.” (For complexity science readers, the notions of transforma-tion through phase transitions is relevant to this discussion).

In other words, “Resilience is the capacity of the affected community to self-organize, learn from and vigorously recover from adverse situations stronger than it was before” (8). This link between resilience and capacity for self-organization is very important, which explains why a recent and major evaluation of the 2010 Haiti Earthquake disaster response promotes the “attainment of self-sufficiency, rather than the ongoing dependency on standard humanitarian assistance.” Indeed, “focus groups indicated that solutions to help people help themselves were desired.”

The fact of the matter is that we are not all affected in the same way during a disaster. (Recall the distinction between hazards and disasters discussed earlier). Those of use who are less affected almost always want to help those in need. Herein lies the critical role of peer-to-peer feedback loops. To be sure, the speed at which the damaged part of an ecosystem can repair itself depends on how many feedback loops it has to the non- (or less-) damaged parts of the eco-system(s). These feedback loops are what enable adaptation and recovery.

Lastly, disaster response professionals cannot be every where at the same time. But the crowd is always there. Moreover, the vast majority of survivals following major disasters cannot be attributed to external aid. One study estimates that at most 10% of external aid contributes to saving lives. Why? Because the real first responders are the disaster-affected communities themselves, the local popula-tion. That is, the real first feedback loops are always local. This dynamic of mutual-aid facilitated by social media is certainly not new, however. My colleagues in Russia did this back in 2010 during the major forest fires that ravaged their country.

While I do have a bias towards people-centered interventions, this does not mean that I discount the importance of feedback loops to external actors such as traditional institutions and humanitarian organizations. I also don’t mean to romanticize the notion of “indigenous technical knowledge” or local coping mechanism. Some violate my own definition of human rights, for example. However, my bias stems from the fact that I am particularly interested in disaster resilience within the context of areas of limited statehood where said institutions and organizations are either absent are ineffective. But I certainly recognize the importance of scale jumping, particularly within the context of social capital and social media.

RESILIENCE THROUGH SOCIAL CAPITAL

Information-based feedback loops general social capital, and the latter has been shown to improve disaster resilience and recovery. In his recent book entitled “Building Resilience: Social Capital in Post-Disaster Recovery,” Daniel Aldrich draws on both qualitative and quantitative evidence to demonstrate that “social resources, at least as much as material ones, prove to be the foundation for resilience and recovery.” His case studies suggest that social capital is more important for disaster resilience than physical and financial capital, and more important than conventional explanations. So the question that naturally follows given our interest in resilience & technology is this: can social media (which is not restricted by geography) influence social capital?

Social Capital

Building on Daniel’s research and my own direct experience in digital humani-tarian response, I argue that social media does indeed nurture social capital during disasters. “By providing norms, information, and trust, denser social networks can implement a faster recovery.” Such norms also evolve on Twitter, as does information sharing and trust building. Indeed, “social ties can serve as informal insurance, providing victims with information, financial help and physical assistance.” This informal insurance, “or mutual assistance involves friends and neighbors providing each other with information, tools, living space, and other help.” Again, this bonding is not limited to offline dynamics but occurs also within and across online social networks. Recall the sand pile analogy. Social capital facilitates the transformation of the sand pile away (temporarily) from self-organized criticality. On a related note vis-a-vis open source software, “the least important part of open source software is the code.” Indeed, more important than the code is the fact that open source fosters social ties, networks, communities and thus social capital.

(Incidentally, social capital generated during disasters is social capital that can subsequently be used to facilitate self-organization for non-violent civil resistance and vice versa).

RESILIENCE through big data

My empirical research on tweets posted during disasters clearly shows that while many use twitter (and social media more generally) to post needs during a crisis, those who are less affected in the social ecosystem will often post offers to help. So where does Big Data fit into this particular equation? When disaster strikes, access to information is equally important as access to food and water. This link between information, disaster response and aid was officially recognized by the Secretary General of the International Federation of Red Cross & Red Crescent Societies in the World Disasters Report published in 2005. Since then, disaster-affected populations have become increasingly digital thanks to the very rapid and widespread adoption of mobile technologies. Indeed, as a result of these mobile technologies, affected populations are increasingly able to source, share and generate a vast amount of information, which is completely transforming disaster response.

In other words, disaster-affected communities are increasingly becoming the source of Big (Crisis) Data during and following major disasters. There were over 20 million tweets posted during Hurricane Sandy. And when the major earth-quake and Tsunami hit Japan in early 2011, over 5,000 tweets were being posted every secondThat is 1.5 million tweets every 5 minutes. So how can Big Data Analytics create more resilience in this respect? More specifically, how can Big Data Analytics accelerate disaster recovery? Manually monitoring millions of tweets per minute is hardly feasible. This explains why I often “joke” that we need a local Match.com for rapid disaster recovery. Thanks to social computing, artifi-cial intelligence, machine learning and Big Data Analytics, we can absolutely develop a “Match.com” for rapid recovery. In fact, I’m working on just such a project with my colleagues at QCRI. We are also developing algorithms to auto-matically identify informative and actionable information shared on Twitter, for example. (Incidentally, a by-product of developing a robust Match.com for disaster response could very well be an increase in social capital).

There are several other ways that advanced computing can create disaster resilience using Big Data. One major challenge is digital humanitarian response is the verification of crowdsourced, user-generated content. Indeed, misinforma-tion and rumors can be highly damaging. If access to information is tantamount to food access as noted by the Red Cross, then misinformation is like poisoned food. But Big Data Analytics has already shed some light on how to develop potential solutions. As it turns out, non-credible disaster information shared on Twitter propagates differently than credible information, which means that the credibility of tweets could be predicted automatically.

Conclusion

In sum, “resilience is the critical link between disaster and development; monitoring it [in real-time] will ensure that relief efforts are supporting, and not eroding […] community capabilities” (9). While the focus of this blog post has been on disaster resilience, I believe the insights provided are equally informa-tive for less extreme events.  So I’d like to end on two major points. The first has to do with data philanthropy while the second emphasizes the critical importance of failing gracefully.

Big Data is Closed and Centralized

A considerable amount of “Big Data” is Big Closed and Centralized Data. Flow-minder’s study mentioned above draws on highly proprietary telecommunica-tions data. Facebook data, which has immense potential for humanitarian response, is also closed. The same is true of Twitter data, unless you have millions of dollars to pay for access to the full Firehose, or even Decahose. While access to the Twitter API is free, the number of tweets that can be downloaded and analyzed is limited to several thousand a day. Contrast this with the 5,000 tweets per second posted after the earthquake and Tsunami in Japan. We therefore need some serious political will from the corporate sector to engage in “data philanthropy”. Data philanthropy involves companies sharing proprietary datasets for social good. Call it Corporate Social Responsibility (CRS) for digital humanitarian response. More here on how this would work.

Failing Gracefully

Lastly, on failure. As noted, complex systems tend towards instability, i.e., self-organized criticality, which is why Homer-Dixon introduces the notion of failing gracefully. “Somehow we have to find the middle ground between dangerous rigidity and catastrophic collapse.” He adds that:

“In our organizations, social and political systems, and individual lives, we need to create the possibility for what computer programmers and disaster planners call ‘graceful’ failure. When a system fails gracefully, damage is limited, and options for recovery are preserved. Also, the part of the system that has been damaged recovers by drawing resources and information from undamaged parts.” Homer-Dixon explains that “breakdown is something that human social systems must go through to adapt successfully to changing conditions over the long term. But if we want to have any control over our direction in breakdown’s aftermath, we must keep breakdown constrained. Reducing as much as we can the force of underlying tectonic stresses helps, as does making our societies more resilient. We have to do other things too, and advance planning for breakdown is undoubtedly the most important.”

As Louis Pasteur famously noted, “Chance favors the prepared mind.” Preparing for breakdown is not defeatist or passive. Quite on the contrary, it is wise and pro-active. Our hubris—including our current infatuation with Bid Data—all too often clouds our better judgment. Like Macbeth, rarely do we seriously ask our-selves what we would do “if we should fail.” The answer “then we fail” is an option. But are we truly prepared to live with the devastating consequences of total synchronous failure?

In closing, some lingering (less rhetorical) questions:

  • How can resilience can be measured? Is there a lowest common denominator? What is the “atom” of resilience?
  • What are the triggers of resilience, creative capacity, local improvisation, regenerative capacity? Can these be monitored?
  • Where do the concepts of “lived reality” and “positive deviance” enter the conversation on resilience?
  • Is resiliency a right? Do we bear a responsibility to render systems more resilient? If so, recalling that resilience is the capacity to self-organize, do local communities have the right to self-organize? And how does this differ from democratic ideals and freedoms?
  • Recent research in social-psychology has demonstrated that mindfulness is an amplifier of resilience for individuals? How can be scaled up? Do cultures and religions play a role here?
  • Collective memory influences resilience. How can this be leveraged to catalyze more regenerative social systems?

bio

Epilogue: Some colleagues have rightfully pointed out that resilience is ultima-tely political. I certainly share that view, which is why this point came up in recent conversations with my PopTech colleagues Andrew Zolli & Leetha Filderman. Readers of my post will also have noted my emphasis on distinguishing between hazards and disasters; that the latter are the product of social, economic and political processes. As noted in my blog post, there are no natural disastersTo this end, some academics rightly warn that “Resilience is a very technical, neutral, apolitical term. It was initially designed to characterize systems, and it doesn’t address power, equity or agency…  Also, strengthening resilience is not free—you can have some winners and some losers.”

As it turns out, I have a lot say about the political versus technical argument. First of all, this is hardly a new or original argument but nevertheless an important one. Amartya Senn discussed this issue within the context of famines decades ago, noting that famines do not take place in democracies. In 1997, Alex de Waal published his seminal book, “Famine Crimes: Politics and the Disaster Relief In-dustry in Africa.” As he rightly notes, “Fighting famine is both a technical and political challenge.” Unfortunately, “one universal tendency stands out: technical solutions are promoted at the expense of political ones.” There is also a tendency to overlook the politics of technical actions, muddle or cover political actions with technical ones, or worse, to use technical measures as an excuse not to undertake needed political action.

De Waal argues that the use of the term “governance” was “an attempt to avoid making the political critique too explicit, and to enable a focus on specific technical aspects of government.” In some evaluations of development and humanitarian projects, “a caveat is sometimes inserted stating that politics lies beyond the scope of this study.” To this end, “there is often a weak call for ‘political will’ to bridge the gap between knowledge of technical measures and action to implement them.” As de Waal rightly notes, “the problem is not a ‘missing link’ but rather an entire political tradition, one manifestation of which is contemporary international humanitarianism.” In sum, “technical ‘solutions’ must be seen in the political context, and politics itself in the light of the domi-nance of a technocratic approach to problems such as famine.”

From a paper I presented back in 2007: “the technological approach almost always serves those who seek control from a distance.” As a result of this technological drive for pole position, a related “concern exists due to the separation of risk evaluation and risk reduction between science and political decision” so that which is inherently politically complex becomes depoliticized and mechanized. In Toward a Rational Society (1970), the German philosopher Jürgen Habermas describes “the colonization of the public sphere through the use of instrumental technical rationality. In this sphere, complex social problems are reduced to technical questions, effectively removing the plurality of contending perspectives.”

To be sure, Western science tends to pose the question “How?” as opposed to “Why?”What happens then is that “early warning systems tend to be largely conceived as hazard-focused, linear, topdown, expert driven systems, with little or no engagement of end-users or their representatives.” As De Waal rightly notes, “the technical sophistication of early warning systems is offset by a major flaw: response cannot be enforced by the populace. The early warning information is not normally made public.”  In other words, disaster prevention requires “not merely identifying causes and testing policy instruments but building a [social and] political movement” since “the framework for response is inherently political, and the task of advocacy for such response cannot be separated from the analytical tasks of warning.”

Recall my emphasis on people-centered early warning above and the definition of resilience as capacity for self-organization. Self-organization is political. Hence my efforts to promote greater linkages between the fields of nonviolent action and early warning years ago. I have a paper (dated 2008) specifically on this topic should anyone care to read. Anyone who has read my doctoral dissertation will also know that I have long been interested in the impact of technology on the balance of power in political contexts. A relevant summary is available here. Now, why did I not include all this in the main body of my blog post? Because this updated section already runs over 1,000 words.

In closing, I disagree with the over-used criticism that resilience is reactive and about returning to initial conditions. Why would we want to be reactive or return to initial conditions if the latter state contributed to the subsequent disaster we are recovering from? When my colleague Andrew Zolli talks about resilience, he talks about “bouncing forward”, not bouncing back. This is also true of Nassim Taleb’s term antifragility, the ability to thrive on disruption. As Homer-Dixon also notes, preparing to fail gracefully is hardly reactive either.

Comparing the Quality of Crisis Tweets Versus 911 Emergency Calls

In 2010, I published this blog post entitled “Calling 911: What Humanitarians Can Learn from 50 Years of Crowdsourcing.” Since then, humanitarian colleagues have become increasingly open to the use of crowdsourcing as a methodology to  both collect and process information during disasters.  I’ve been studying the use of twitter in crisis situations and have been particularly interested in the quality, actionability and credibility of such tweets. My findings, however, ought to be placed in context and compared to other, more traditional, reporting channels, such as the use of official emergency telephone numbers. Indeed, “Information that is shared over 9-1-1 dispatch is all unverified information” (1).

911ex

So I did some digging and found the following statistics on 911 (US) & 999 (UK) emergency calls:

  • “An astounding 38% of some 10.4 million calls to 911 [in New York City] during 2010 involved such accidental or false alarm ‘short calls’ of 19 seconds or less — that’s an average of 10,700 false calls a day”.  – Daily News
  • “Last year, seven and a half million emergency calls were made to the police in Britain. But fewer than a quarter of them turned out to be real emergencies, and many were pranks or fakes. Some were just plain stupid.” – ABC News

I also came across the table below in this official report (PDF) published in 2011 by the European Emergency Number Association (EENA). The Greeks top the chart with a staggering 99% of all emergency calls turning out to be false/hoaxes, while Estonians appear to be holier than the Pope with less than 1% of such calls.

Screen Shot 2012-12-11 at 4.45.34 PM

Point being: despite these “data quality” issues, European law enforcement agencies have not abandoned the use of emergency phone numbers to crowd-source the reporting of emergencies. They are managing the challenge since the benefit of these number still far outweigh the costs. This calculus is unlikely to change as law enforcement agencies shift towards more mobile-based solutions like the use of SMS for 911 in the US. This important shift may explain why tra-ditional emergency response outfits—such as London’s Fire Brigade—are putting in place processes that will enable the public to report via Twitter.

For more information on the verification of crowdsourced social media informa-tion for disaster response, please follow this link.

What Waze Can Teach Us About Crowdsourcing and Crisis Mapping

I recently tried out Waze, “the world’s fastest-growing community-based traffic and navigation app.” The smart phone app enables drivers to easily share real-time traffic and road information, “saving everyone time and gas money on their daily commute.” Waze is quite possibly the most successful live, crowdsourced mapping platform there is. Over 30 million drivers use the app to “outsmart traffic and get everyone the best route to work and back, every day.”

Why is Waze so successful? Because the creators of this app recognize full well that “Traffic is more than just red lines on the map.” Likewise, Crisis Mapping is also more than just dots on the map. But unlike Waze, groups like Ushahidi have not adopted dynamic geo-fencing features (despite my best efforts back in 2011). With Waze, drivers get automatically alerted before they approach police, accidents, road hazards or traffic jams, all shared by other drivers in real time. “It’s like a personal heads-up from a few million of your friends on the road.” The creators of Waze have also integrated gaming and social networking features into their app, something I also lobbied Ushahidi to do years ago. Not only is updating Waze super easy and fast to update, it is also fun and rewarding—which reminds me of the “Fisher Price Theory of Crisis Mapping.”

Just like drivers on the motorway, humanitarians do not have the time to keep watching and analyzing dots on a map. They have to keep their hands “on the wheel” and focus on the more important tasks at hand. Crisis Mapping platforms therefore have to be hands-free and more voice-based to limit the distraction of tactile data entry. In other words, the interface needs to become invisible. As computing pioneer Mark Weiser noted, “The best technology should be invisible, get out of your way, and let you live your life.” My colleague Amber Case add that “We shouldn’t have to fiddle with interfaces. We should be humans; machines should be machines; each amplifying the best of both. Wouldn’t that make for a nice reality?” This is what Waze is doing by integrating very neat user-interface features and voice-based controls, which improve situational awareness and facilitates real-time decision-making.

In conclusion, Waze’s clever user-centered design features are also relevant to map-based development and human rights projects in the majority world (i.e. developing countries). Otherwise, the value of digital maps is more like that of news articles—that is, informative but not necessarily operational and actionable for local decision-making purposes.

bio

See also: Waze Helps Drivers Get Around Damage from Oklahoma Tornadoes

Tweeting is Believing? Analyzing Perceptions of Credibility on Twitter

What factors influence whether or not a tweet is perceived as credible? According to this recent study, users have “difficulty discerning truthfulness based on con-tent alone, with message topic, user name, and user image all impacting judg-ments of tweets and authors to varying degrees regardless of the actual truth-fulness of the item.”

For example, “Features associated with low credibility perceptions were the use of non-standard grammar and punctuation, not replacing the default account image, or using a cartoon or avatar as an account image. Following a large number of users was also associated with lower author credibility, especially when unbalanced in comparison to follower count […].” As for features enhan-cing a tweet’s credibility, these included “author influence (as measured by follower, retweet, and  mention counts), topical expertise (as established through a Twitter homepage bio, history of on-topic tweeting, pages outside of Twitter, or having a location relevant to the topic of the tweet), and reputation (whether an author is someone a user follows, has heard of, or who has an official Twitter account verification seal). Content related features viewed as credibility-enhancing were containing a URL leading to a high-quality site, and the existence of other tweets conveying similar information.”

 In general, users’ ability to “judge credibility in practice is largely limited to those features visible at-a-glance in current UIs (user picture, user name, and tweet content). Conversely, features that often are obscured in the user interface, such as the bio of a user, receive little attention despite their ability to impact cred-ibility judgments.” The table below compares a features’s perceived credibility impact with the attention actually allotted to assessing that feature.

“Message topic influenced perceptions of tweet credibility, with science tweets receiving a higher mean tweet credibility rating than those about either politics  or entertainment. Message topic had no statistically significant impact on perceptions of author credibility.” In terms of usernames, “Authors with topical names were considered more credible than those with traditional user names, who were in turn considered more credible than those with internet name styles.” In a follow up experiment, the study analyzed perceptions of credibility vis-a-vis a user’s image, i.e., the profile picture associated with a given Twitter account. “Use of the default Twitter icon significantly lowers ratings of content and marginally lowers ratings of authors […]” in comparison to generic, topical, female and male images.

Obviously, “many of these metrics can be faked to varying extents. Selecting a topical username is trivial for a spam account. Manufacturing a high follower to following ratio or a high number of retweets is more difficult but not impossible. User interface changes that highlight harder to fake factors, such as showing any available relationship between a user’s network and the content in question, should help.” Overall, these results “indicate a discrepancy between features people rate as relevant to determining credibility and those that mainstream social search engines make available.” The authors of the study conclude by suggesting changes in interface design that will enhance a user’s ability to make credibility judgements.

“Firstly, author credentials should be accessible at a glance, since these add value and users rarely take the time to click through to them. Ideally this will include metrics that convey consistency (number of tweets on topic) and legitimization by other users (number of mentions or retweets), as well as details from the author’s Twitter page (bio, location, follower/following counts). Second, for con-tent assessment, metrics on number of retweets or number of times a link has been shared, along with who is retweeting and sharing, will provide consumers with context for assessing credibility. […] seeing clusters of tweets that conveyed similar messages was reassuring to users; displaying such similar clusters runs counter to the current tendency for search engines to strive for high recall by showing a diverse array of retrieved items rather than many similar ones–exploring how to resolve this tension is an interesting area for future work.”

In sum, the above findings and recommendations explain why platforms such as RapportiveSeriously Rapid Source Review (SRSR) and CrisisTracker add so much value to the process of assessing the credibility of tweets in near real-time. For related research: Predicting the Credibility of Disaster Tweets Automatically and: Automatically Ranking the Credibility of Tweets During Major Events.

How Can Digital Humanitarians Best Organize for Disaster Response?

My colleague Duncan Watts recently spoke with Scientific American about a  new project I am collaborating on with him & colleagues at Microsoft Research. I first met Duncan while at the Santa Fe Institute (SFI) back in 2006. We recently crossed paths again (at 10 Downing Street, of all places), and struck up a conver-sation about crisis mapping and the Standby Volunteer Task Force (SBTF). So I shared with him some of the challenges we were facing vis-a-vis the scaling up of our information processing workflows for digital humanitarian response. Duncan expressed a strong interest in working together to address some of these issues. As he told Scientific American, “We’d like to help them by trying to understand in a more scientific manner how to scale up information processing organizations like the SBTF without over-loading any part of the system.”

Scientific American Title

Here are the most relevant sections of his extended interview:

In addition to improving research methods, how might the Web be used to deliver timely, meaningful research results?

Recently, a handful of volunteer “crisis mapping” organizations such as The Standby Task Force [SBTF] have begun to make a difference in crisis situations by performing real-time monitoring of information sources such as Facebook, Twitter and other social media, news reports and so on and then superposing these reports on a map interface, which then can be used by relief agencies and affected populations alike to improve their under-standing of the situation. Their efforts are truly inspiring, and they have learned a lot from experience. We want to build off that real-world model through Web-based crisis-response drills that test the best ways to comm-unicate and coordinate resources during and after a disaster.

How might you improve upon existing crisis-mapping efforts?

The efforts of these crisis mappers are truly inspiring, and groups like the SBTF have learned a lot about how to operate more effectively, most from hard-won experience.  At the same time, they’ve encountered some limita-tions to their model, which depends critically on a relatively small number of dedicated individuals, who can easily get overwhelmed or burned out. We’d like to help them by trying to understand in a more scientific manner how to scale up information processing organizations like the SBTF without over-loading any part of the system.

How would you do this in the kind of virtual lab environment you’ve been describing?

The basic idea is to put groups of subjects into simulated crisis-mapping drills, systematically vary different ways of organizing them, and measure how quickly and accurately they collectively process the corresponding information. So for any given drill, the organizer would create a particular disaster scenario, including downed power lines, fallen trees, fires and flooded streets and homes. The simulation would then generate a flow of information, like a live tweet stream that resembles the kind of on-the-ground reporting that occurs in real events, but in a controllable way.

As a participant in this drill, imagine you’re monitoring a Twitter feed, or some other stream of reports, and that your job is to try to accurately recreate the organizer’s disaster map based on what you’re reading. So for example, you’re looking at Twitter feeds for everything during hurricane Sandy that has “#sandy” associated with it. From that information, you want to build a map of New York and the tri-state region that shows everywhere there’s been lost power, everywhere there’s a downed tree, everywhere where there’s a fire.

You could of course try to do this on your own, but as the rate of infor-mation flow increased, any one person would get overwhelmed; so it would be necessary to have a group of people working on it together. But depen-ding on how the group is organized, you could imagine that they’d do a better or worse job, collectively. The goal of the experiment then would be to measure the performance of different types of organizations—say with different divisions of labor or different hierarchies of management—and discover which work better as a function of the complexity of the scenario you’ve presented and the rate of information being generated. This is something that we’re trying to build right now.

What’s the time frame for implementing such crowdsourced disaster mapping drills?

We’re months away from doing something like this. We still need to set up the logistics and are talking to a colleague [Patrick Meier] who works as a crisis mapper to get a better understanding of how they do things so that we can design the experiment in a way that is motivated by a real problem.

How will you know when your experiments have created something valuable for better managing disaster responses?

There’s no theory that says, here’s the best way to organize n people to process the maximum amount of information reliably. So ideally we would like to design an experiment that is close enough to realistic crisis-mapping scenarios that it could yield some actionable insights. But the experiment would also need to be sufficiently simple and abstract so that we learn something about how groups of people process information that generalizes beyond the very specific case of crisis mapping.

As a scientist, I want to identify causal mechanisms in a nice, clean way and reduce the problem to its essence. But as someone who cares about making a difference in the real world, I would also like to be able to go back to my friend who’s a crisis mapper and say we did the experiment, and here’s what the science says you should do to be more effective.

The full interview is available at Scientific AmericanStay tuned for further up-dates on this research.

How the UN Used Social Media in Response to Typhoon Pablo (Updated)

Our mission as digital humanitarians was to deliver a detailed dataset of pictures and videos (posted on Twitter) which depict damage and flooding following the Typhoon. An overview of this digital response is available here. The task of our United Nations colleagues at the Office of the Coordination of Humanitarian Affairs (OCHA), was to rapidly consolidate and analyze our data to compile a customized Situation Report for OCHA’s team in the Philippines. The maps, charts and figures below are taken from this official report (click to enlarge).

Typhon PABLO_Social_Media_Mapping-OCHA_A4_Portrait_6Dec2012

This map is the first ever official UN crisis map entirely based on data collected from social media. Note the “Map data sources” at the bottom left of the map: “The Digital Humanitarian Network’s Solution Team: Standby Volunteer Task Force (SBTF) and Humanity Road (HR).” In addition to several UN agencies, the government of the Philippines has also made use of this information.

Screen Shot 2012-12-08 at 7.26.19 AM

Screen Shot 2012-12-08 at 7.29.24 AM

The cleaned data was subsequently added to this Google Map and also made public on the official Google Crisis Map of the Philippines.

Screen Shot 2012-12-08 at 7.32.17 AM

One of my main priorities now is to make sure we do a far better job at leveraging advanced computing and microtasking platforms so that we are better prepared the next time we’re asked to repeat this kind of deployment. On the advanced computing side, it should be perfectly feasible to develop an automated way to crawl twitter and identify links to images  and videos. My colleagues at QCRI are already looking into this. As for microtasking, I am collaborating with PyBossa and Crowdflower to ensure that we have highly customizable platforms on stand-by so we can immediately upload the results of QCRI’s algorithms. In sum, we have got to move beyond simple crowdsourcing and adopt more agile micro-tasking and social computing platforms as both are far more scalable.

In the meantime, a big big thanks once again to all our digital volunteers who made this entire effort possible and highly insightful.

Statistics on First Tweets to Report the #Japan Earthquake (Updated)

Update: The first (?) YouTube video of earthquake shared on Twitter.

An 7.3 magnitude earthquake just struck 300km off the eastern coast of Japan, prompting a tsunami warning for Japan’s Miyagi Prefecture. The quake struck at 5.18pm local time (3.18am New York Time). Twitter’s team in Japan have just launched this page of recommended hashtags. There are currently over 1,200 tweets per minute being posted in Tokyo, according to this site.

Screen Shot 2012-12-07 at 4.20.49 AM

Hashtags.org has the following graph on the frequency of tweets carrying the Japan #hashtag over the past 24 hours:

Screen Shot 2012-12-07 at 4.27.52 AM

The first tweets to report the earthquake on twitter using the hashtag #Japan were posted at 5.19pm local time (3.19am New York). You can click on each for the original link.

Screen Shot 2012-12-07 at 4.07.43 AM

Screen Shot 2012-12-07 at 4.08.05 AM Screen Shot 2012-12-07 at 4.08.20 AM

Screen Shot 2012-12-07 at 4.17.53 AM

Screen Shot 2012-12-07 at 4.16.11 AM

 Screen Shot 2012-12-07 at 4.10.35 AM Screen Shot 2012-12-07 at 4.10.55 AM Screen Shot 2012-12-07 at 4.11.16 AM

These tweets were each posted within 2 minutes of the earthquake. I will update this blog post when I get more relevant details.

Summary: Digital Disaster Response to Philippine Typhoon

Update: How the UN Used Social Media in Response to Typhoon Pablo

The United Nations Office for the Coordination of Humanitarian Affairs (OCHA) activated the Digital Humanitarian Network (DHN) on December 5th at 3pm Geneva time (9am New York). The activation request? To collect all relevant tweets about Typhoon Pablo posted on December 4th and 5th; identify pictures and videos of damage/flooding shared in those tweets; geo-locate, time-stamp and categorize this content. The UN requested that this database be shared with them by 5am Geneva time the following day. As per DHN protocol, the activation request was reviewed within an hour. The UN was informed that the request had been granted and that the DHN was formally activated at 4pm Geneva.

pablo_impact

The DHN is composed of several members who form Solution Teams when the network is activated. The purpose of Digital Humanitarians is to support humanitarian organizations in their disaster response efforts around the world. Given the nature of the UN’s request, both the Standby Volunteer Task Force (SBTF) and Humanity Road (HR) joined the Solution Team. HR focused on analyzing all tweets posted December 4th while the SBTF worked on tweets posted December 5th. Over 20,000 tweets were analyzed. As HR will have a blog post describing their efforts shortly (please check here), I will focus on the SBTF.

Geofeedia Pablo

The Task Force first used Geofeedia to identify all relevant pictures/videos that were already geo-tagged by users. About a dozen were identified in this manner. Meanwhile, the SBTF partnered with the Qatar Foundation Computing Research Institute’s (QCRI) Crisis Computing Team to collect all tweets posted on December 5th with the hashtags endorsed by the Philippine Government. QCRI ran algorithms on the dataset to remove (1) all retweets and (2) all tweets without links (URLs). Given the very short turn-around time requested by the UN, the SBTF & QCRI Teams elected to take a two-pronged approach in the hopes that one, at least, would be successful.

The first approach used  Crowdflower (CF), introduced here. Workers on Crowd-flower were asked to check each Tweet’s URL and determine whether it linked to a picture or video. The purpose was to filter out URLs that linked to news articles. CF workers were also asked to assess whether the tweets (or pictures/videos) provided sufficient geographic information for them to be mapped. This methodology worked for about 2/3 of all the tweets in the database. A review of lessons learned and how to use Crowdflower for disaster response will be posted in the future.

Pybossa Philippines

The second approach was made possible thanks to a partnership with PyBossa, a free, open-source crowdsourcing and micro-tasking platform. This effort is described here in more detail. While we are still reviewing the results of this approach, we expect that  this tool will become the standard for future activations of the Digital Humanitarian Network. I will thus continue working closely with the PyBossa team to set up a standby PyBossa platform ready-for-use at a moment’s notice so that Digital Humanitarians can be fully prepared for the next activation.

Now for the results of the activation. Within 10 hours, over 20,000 tweets were analyzed using a mix of methodologies. By 4.30am Geneva time, the combined efforts of HR and the SBTF resulted in a database of 138 highly annotated tweets. The following meta-data was collected for each tweet:

  • Media Type (Photo or Video)
  • Type of Damage (e.g., large-scale housing damage)
  • Analysis of Damage (e.g., 5 houses flooded, 1 damaged roof)
  • GPS coordinates (latitude/longitude)
  • Province
  • Region
  • Date
  • Link to Photo or Video

The vast majority of curated tweets had latitude and longitude coordinates. One SBTF volunteer (“Mapster”) created this map below to plot the data collected. Another Mapster created a similar map, which is available here.

Pablo Crisis Map Twitter Multimedia

The completed database was shared with UN OCHA at 4.55am Geneva time. Our humanitarian colleagues are now in the process of analyzing the data collected and writing up a final report, which they will share with OCHA Philippines today by 5pm Geneva time.

Needless to say, we all learned a lot thanks to the deployment of the Digital Humanitarian Network in the Philippines. This was the first time we were activated to carry out a task of this type. We are now actively reviewing our combined efforts with the concerted aim of streamlining our workflows and methodologies to make this type effort far easier and quicker to complete in the future. If you have suggestions and/or technologies that could facilitate this kind of digital humanitarian work, then please do get in touch either by posting your ideas in the comments section below or by sending me an email.

Lastly, but definitely most importantly, a big HUGE thanks to everyone who volunteered their time to support the UN’s disaster response efforts in the Philippines at such short notice! We want to publicly recognize everyone who came to the rescue, so here’s a list of volunteers who contributed their time (more to be added!). Without you, there would be no database to share with the UN, no learning, no innovating and no demonstration that digital volunteers can and do make a difference. Thank you for caring. Thank you for daring.

Help Tag Tweets from Typhoon Pablo to Support UN Disaster Response!

Update: Summary of digital humanitarian response efforts available here.

The United Nations Office for the Coordination of Humanitarian Affairs (OCHA) has just activated the Digital Humanitarian Network (DHN) to request support in response to Typhoo Pablo. They also need your help! Read on!

pablopic

The UN has asked for pictures and videos of the damage to be collected from tweets posted over the past 48 hours. These pictures/videos need to be geo-tagged if at all possible, and time-stamped. The Standby Volunteer Task Force (SBTF) and Humanity Road (HR), both members of Digital Humanitarians, are thus collaborating to provide the UN with the requested data, which needs to be submitted by today 10pm 11pm New York time, 5am Geneva time tomorrow. Given this very short turn around time, we only have 10 hours (!), the Digital Humani-tarian Network needs your help!

Pybossa Philippines

The SBTF has partnered with colleagues at PyBossa to launch this very useful microtasking platform for you to assist the UN in these efforts. No prior experience necessary. Click here or on the display above to see just how easy it is to support the disaster relief operations on the ground.

A very big thanks to Daniel Lombraña González from PyBossa for turning this around at such short notice! If you have any questions about this project or with respect to volunteering, please feel free to add a comment to this blog post below. Even if you only have time tag one tweet, it counts! Please help!

Some background information on this project is available here.

Analyzing Disaster Tweets from Major Thai Floods

The 2011 Thai Floods was one of the country’s worst disasters in recent history.  The flooding began in July and lasted until December. Over 13 million people were affected. More than 800 were killed. The World Bank estimated $45 billion in total economic damage. This new study, “The Role of Twitter during a Natural Disaster: Case Study of 2011 Thai Flood,” analyzes how twitter was used during these major floods.

The number of tweets increase significantly in October, which is when the flooding reached parts of the Bangkok Metropolitan area. The month before (Sept-to-Oct) also a notable increase of tweets, which may “demonstrate that Thais were using Twitter to search for realtime and practical information that traditional media could not provide during the natural disaster period.”

To better understand the type of information shared on Twitter during the floods, the authors analyzed 175,551 tweets that used the hashtag #thaiflood. They removed “retweets” and duplicates, yielding a dataset of 64,582 unique tweets. Using keyword analysis and a rule based approach, the authors auto-matically classified these tweets into 5 categories:

Situational Announcements and Alerts: Tweets about up-to-date situational and location-based information related to the flood such as water levels, traffic conditions and road conditions in certain areas. In addition, emergency warnings from authorities advising citizens to evacuate areas, seek shelter or take other protective measures are also included.

Support Announcements: Tweets about free parking availability, free emergency survival kits distribution and free consulting services for home repair, etc.

Requests for Assistance: Tweets requesting any types of assistance; such as food, water, medical supplies, volunteers or transportation.

Requests for Information: Tweets including general inquiries related to the flood and flood relief such as inquiries for telephone numbers of relevant authorities, regarding the current situation in specific locations and about flood damage compensation.

Other: Tweets including all other messages, such as general comments; complaints and expressions of opinions.

The results of this analysis are shown in the figures below. The first shows the number of tweets per each category, while the second shows the distribution of these categories over time.

Messages posted during the first few weeks “included current water levels in certain areas and roads; announcements for free parking availability; requests for volunteers to make sandbags and pack emergency survival kits; announce-ments for evacuation in certain areas and requests for boats, food, water supplies and flood donation information. For the last few weeks when water started to recede, Tweet messages included reports on areas where water had receded, information on home cleaning andrepair and guidance regarding the process to receive flood damage compensation from the government.”

To determine the credibility of tweets, the authors identify the top 10 most re-tweeted users during the floods. They infer that the most retweeted tweets signal that the content of said tweets is perceived as credible. “The majority of these top users are flood/disaster related government or private organizations.” Siam Arsa, one of the leading volunteer networks helping flood victims in Thailand, was one of the top users ranked by retweets. The group utilizes social media on both Facebook  (www.facebook.com/siamarsa) and Twitter (@siamarsa) to share information about flooding and related volunteer work.”

In conclusion, “if the government plans to implement social media as a tool for disaster response, it would be well advised to prepare some measures or pro-tocols that help officials verify incoming information and eliminate false information. The  citizens should also be educated to take caution when receiving news and information via social media, and to think carefully about the potential effect before disseminating certain content.”

Gov Twitter

My QCRI colleagues and I are collecting tweets about Typhoon Pablo, which is making landfall in the Philippines. We’re specifically tracking tweets with one or more of the following hashtags: #PabloPh, #reliefPH and #rescuePH, which the government is publicly encouraging Filipinos to use. We hope to carry out an early analysis of these tweets to determine which ones provide situational aware-ness. The purpose of this applied action research is to ultimately develop a real-time dashboard for humanitarian response. This explains why we launched this Library of Crisis Hashtags. For further reading, please see this post on “What Percentage of Tweets Generated During a Crisis Are Relevant for Humanitarian Response?”