Combining Satellite Imagery and Machine Learning to Predict Poverty

Curated on Posted on by Stefaan Verhulst

From the sustainability and artificial intelligence lab: “The elimination of poverty worldwide is the first of 17 UN Sustainable Development Goals for the year 2030. To track progress towards this goal, we require more frequent and more reliable data on the distribution of poverty than traditional data collection methods can provide.

In this project, we propose an approach that combines machine learning with high-resolution satellite imagery to provide new data on socioeconomic indicators of poverty and wealth. Check out the short video below for a quick overview and then read the paper for a more detailed explanation of how it all works….(More)”

National Transit Map Seeks to Close the Transit Data Gap

Curated on Posted on by Stefaan Verhulst

Ben Miller at GovTech: “In bringing together the first ever map illustrating the nation’s transit system, the U.S. Department of Transportation isn’t just making data more accessible — it’s also aiming to modernize data collection and dissemination for many of the country’s transit agencies.

With more than 10,000 routes and 98,000 stops represented, the National Transit Map is already enormous. But Dan Morgan, chief data officer of the department, says it’s not enough. When measuring vehicles operated in maximum service — a metric illustrating peak service at a transit agency — the National Transit Map captures only about half of all transit in the U.S.

“Not all of these transit agencies have this data available,” Morgan said, “so this is an ongoing project to really close the transit data gap.”Which is why, in the process of building out the map, the DOT is working with transit agencies to make their data available.

Which is why, in the process of building out the map, the DOT is working with transit agencies to make their data available.

On the whole, transit data is easier to collect and process than a lot of transportation data because many agencies have adopted a standard called General Transit Feed Specification (GTFS) that applies to schedule-related data. That’s what made the National Transit Map an easy candidate for completion, Morgan said.

But as popular as GTFS has become, many agencies — especially smaller ones — haven’t been able to use it. The tools to convert to GTFS come with a learning curve.

“It’s really a matter of priority and availability of resources,” he said.

Bringing those agencies into the mainstream is important to achieving the goals of the map. In the map, Morgan said he sees an opportunity to achieve a new level of clarity where it has never existed before.

That’s because transit has long suffered from difficulty in seeing its own history. Transit officials can describe their systems as they exist, but looking at how they got there is trickier.

“There’s no archive,” Morgan said, “there’s no picture of how transit changes over time.”

And that’s a problem for assessing what works and what doesn’t, for understanding why the system operates the way it does and how it responds to changes. …(More)”

More African governments are enacting open data policies but still aren’t willing to share information

Curated on Posted on by Stefaan Verhulst

Joshua Masinde at Quartz Africa: “Working as a data journalist and researcher in Uganda, Lydia Namubiru does not remember a moment she had an easy time accessing official government data in the execution of her work. She has had to literally beg for such information from officials with little success.

In June this year, she approached the Uganda Bureau of Statistics seeking a nationally representative sample of micro data from the country’s 2014 census. Despite frequent calls and emails, she is still waiting for the information from the bureau several months down the line….

It doesn’t have to be that way of course. In neighboring Kenya there’s much optimism there’ll be a different attitude to open data. Last month civil society activists and supporters of open data celebrated the government signing the Access to Information bill into law. It comes after many years of lobbying….

Despite well-earned reputations of authoritarianism and conservative attitudes to governance, it turns out more African governments are opening up to their citizens in the guise of espousing transparency and accountability in the conduct of their affairs.

However, in truth, a government saying it’s allowing citizens to access data or information is very different from the actual practice of enabling that access. For the most part, several governments’ open data initiatives often serve far more mundane purposes and may not be the data that citizens really want—the kind that potentially exposes corruption or laxity in public service…

“Countries that have embraced open data have seen real savings in public spending and improved efficiency in services. Nowhere is this more vital than in our nations – many of which face severe health and education crises,” Nnenna Nwakanma, Africa regional coordinator at World Wide Web Foundation,points out.

 What is more prevalent now is what some open data advocates call ‘open washing’, which is described as a real threat to the open data movement according to the World Wide Web Foundation. By ‘open washing’, governments merely enact open data policies but do not follow through to full implementation. Others simply put in place strong freedom of information and right to information laws but do not really let the citizens take full advantage of the open data. This could, however, be as a result of institutional shortcomings, internal bureaucracies or lack of political will.

As the initiatives towards open data gather steam, challenges such as government agencies being unwilling to release official information as well as state bureaucracies are still prominent. Many governments are also only keen on releasing information that will not portray them as ‘naked’ but that which they feel will project them in positive light. But, as to whether laws will make governments more open, even with the information that citizens really need, is a matter of conjecture. For Namubiru, open data should be a culture that grows more subtly than by way of just passing laws for the sake of it.

“If they release enough packets of data on what they consider neutral or positive information, the storytellers will still be able to connect the dots.”…(More)”

Recent Developments in Open Data Policy

Curated on Posted on by Stefaan Verhulst

Presentation by Paul Uhlir:  “Several International organizations have issued policy statements on open data policies in the past two years. This presentation provides an overview of those statements and their relevance to developing countries.

International Statements on Open Data Policy

Open data policies have become much more supported internationally in recent years. Policy statements in just the most recent 2014-2016 period that endorse and promote openness to research data derived from public funding include: the African Data Consensus (UNECA 2014); the CODATA Nairobi Principles for Data Sharing for Science and Development in Developing Countries (PASTD 2014); the Hague Declaration on Knowledge Discovery in the Digital Age (LIBER 2014); Policy Guidelines for Open Access and Data Dissemination and Preservation (RECODE 2015); Accord on Open Data in a Big Data World (Science International 2015). This presentation will present the principal guidelines of these policy statements.

The Relevance of Open Data from Publicly Funded Research for Development

There are many reasons that publicly funded research data should be made as freely and openly available as possible. Some of these are noted here, although many other benefits are possible. For research, it is closing the gap with more economically developed countries, making researchers more visible on the web, enhancing their collaborative potential, and linking them globally. For educational benefits, open data assists greatly in helping students learn how to do data science and to manage data better. From a socioeconomic standpoint, open data policies have been shown to enhance economic opportunities and to enable citizens to improve their lives in myriad ways. Such policies are more ethical in allowing access to those that have no means to pay and not having to pay for the data twice—once through taxes to create the data in the first place and again at the user level . Finally, access to factual data can improve governance, leading to better decision making by policymakers, improved oversight by constituents, and digital repatriation of objects held by former colonial powers.

Some of these benefits are cited directly in the policy statements themselves, while others are developed more fully in other documents (Bailey Mathae and Uhlir 2012, Uhlir 2015). Of course, not all publicly funded data and information can be made available and there are appropriate reasons—such as the protection of national security, personal privacy, commercial concerns, and confidentiality of all kinds—that make the withholding of them legal and ethical. However, the default rule should be one of openness, balanced against a legitimate reason not to make the data public….(More)”

How Citizen Attachment to Neighborhoods Helps to Improve Municipal Services and Public Spaces

Curated on Posted on by Stefaan Verhulst

Paper by Daniel O’Brien, Dietmar Offenhuber, Jessica Baldwin-Philippi, Melissa Sands, and Eric Gordon: “What motivates people to contact their local governments with reports about street light outages, potholes, graffiti, and other deteriorations in public spaces? Current efforts to improve government interactions with constituents operate on the premise that citizens who make such reports are motivated by broad civic values. In contrast, our recent research demonstrates that such citizens are primarily motivated by territoriality – that is, attachments to the spaces where they live. Our research focuses on Boston’s “311 system,” which provides telephone hotlines and web channels through which constituents can request non-emergency government services.

Although our study focuses on 311 users in Boston, it holds broader implications for more than 400 U.S. municipalities that administer similar systems. And our results encourage a closer look at the drivers of citizen participation in many “coproduction programs” – programs that involve people in the design and implementation of government services. Currently, 311 is just one example of government efforts to use technology to involve constituents in joint efforts.

Territorial Ties and Civic Engagement

The concept of territoriality originated in studies of animal behavior – such as bears marking trees in the forest or lions and hyenas fighting over a kill. Human beings also need to manage the ownership of objects and spaces, but social psychologists have demonstrated that human territoriality, whether at home, the workplace, or a neighborhood, entails more than the defense of objects or spaces against others. It includes maintenance and caretaking, and even extends to items shared with others….(More)”

Artificial intelligence is hard to see

Curated on Posted on by Stefaan Verhulst

Kate Crawford and Meredith Whittaker on “Why we urgently need to measure AI’s societal impacts“: “How will artificial intelligence systems change the way we live? This is a tough question: on one hand, AI tools are producing compelling advances in complex tasks, with dramatic improvements in energy consumption, audio processing, and leukemia detection. There is extraordinary potential to do much more in the future. On the other hand, AI systems are already making problematic judgements that are producing significant social, cultural, and economic impacts in people’s everyday lives.

AI and decision-support systems are embedded in a wide array of social institutions, from influencing who is released from jail to shaping the news we see. For example, Facebook’s automated content editing system recently censored the Pulitzer-prize winning image of a nine-year old girl fleeing napalm bombs during the Vietnam War. The girl is naked; to an image processing algorithm, this might appear as a simple violation of the policy against child nudity. But to human eyes, Nick Ut’s photograph, “The Terror of War”, means much more: it is an iconic portrait of the indiscriminate horror of conflict, and it has an assured place in the history of photography and international politics. The removal of the image caused an international outcry before Facebook backed down and restored the image. “What they do by removing such images, no matter what good intentions, is to redact our shared history,” said the Prime Minister of Norway, Erna Solberg.

It’s easy to forget that these high-profile instances are actually the easy cases. As Tarleton Gillespie has observed, hundreds of content reviews are occurring with Facebook images thousand of times per day, and rarely is there a Pulitzer prize to help determine lasting significance. Some of these reviews include human teams, and some do not. In this case, there is alsoconsiderable ambiguity about where the automated process ended and the human review began: which is part of the problem. And Facebook is just one player in complex ecology of algorithmically-supplemented determinations with little external monitoring to see how decisions are made or what the effects might be.

The ‘Terror of War’ case, then, is the tip of the iceberg: a rare visible instance that points to a much larger mass of unseen automated and semi-automated decisions. The concern is that most of these ‘weak AI’ systems are making decisions that don’t garner such attention. They are embedded at the back-end of systems, working at the seams of multiple data sets, with no consumer-facing interface. Their operations are mainly unknown, unseen, and with impacts that take enormous effort to detect.

Sometimes AI techniques get it right, and sometimes they get it wrong. Only rarely will those errors be seen by the public: like the Vietnam war photograph, or when a AI ‘beauty contest’ held this month was called out for being racist for selecting white women as the winners. We can dismiss this latter case as a problem of training data — they simply need a more diverse selection of faces to train their algorithm with, and now that 600,000 people have sent in their selfies, they certainly have better means to do so. But while a beauty contest might seem like a bad joke, or just a really good trick to get people to give up their photos to build a large training data set, it points to a much bigger set of problems. AI and decision-support systems are reaching into everyday life: determining who will be on a predictive policing‘heat list’, who will be hired or promoted, which students will be recruited to universities, or seeking to predict at birth who will become a criminal by the age of 18. So the stakes are high…(More)”

Law in the Future

Curated on Posted on by Stefaan Verhulst

Paper by Benjamin Alarie, Anthony Niblett and Albert Yoon: “The set of tasks and activities in which humans are strictly superior to computers is becoming vanishingly small. Machines today are not only performing mechanical or manual tasks once performed by humans, they are also performing thinking tasks, where it was long believed that human judgment was indispensable. From self-driving cars to self-flying planes; and from robots performing surgery on a pig to artificially intelligent personal assistants, so much of what was once unimaginable is now reality. But this is just the beginning of the big data and artificial intelligence revolution. Technology continues to improve at an exponential rate. How will the big data and artificial intelligence revolutions affect law? We hypothesize that the growth of big data, artificial intelligence, and machine learning will have important effects that will fundamentally change the way law is made, learned, followed, and practiced. It will have an impact on all facets of the law, from the production of micro-directives to the way citizens learn of their legal obligations. These changes will present significant challenges to human lawmakers, judges, and lawyers. While we do not attempt to address all these challenges, we offer a short and positive preview of the future of law: a world of self-driving law, of legal singularity, and of the democratization of the law…(More)”

Evidence-based policy making in the social sciences: Methods that matter

Curated on Posted on by Stefaan Verhulst

Book edited by Gerry Stoker and Mark Evans: “Drawing on the insights of some of the world’s leading authorities in public policy analysis, this important book offers a distinct and critical showcase of emerging forms of discovery for policy-making. Chapter by chapter this expert group of social scientists showcase their chosen method or approach, showing the context, the method’s key features and how it can be applied in practice, including the scope and limitations of its application and value to policy makers. Arguing that it is not just econometric analysis, cost benefit or surveys that can do policy work, the contributors demonstrate a range of other methods that can provide evidenced-based policy insights and how they can help facilitate progressive policy outcomes…(More)”

A cautionary tale about humans creating biased AI models

Curated on Posted on by Stefaan Verhulst

 at TechCrunch: “Most artificial intelligence models are built and trained by humans, and therefore have the potential to learn, perpetuate and massively scale the human trainers’ biases. This is the word of warning put forth in two illuminating articles published earlier this year by Jack Clark at Bloomberg and Kate Crawford at The New York Times.

Tl;dr: The AI field lacks diversity — even more spectacularly than most of our software industry. When an AI practitioner builds a data set on which to train his or her algorithm, it is likely that the data set will only represent one worldview: the practitioner’s. The resulting AImodel demonstrates a non-diverse “intelligence” at best, and a biased or even offensive one at worst….

So what happens when you don’t consider carefully who is annotating the data? What happens when you don’t account for the differing preferences, tendencies and biases among varying humans? We ran a fun experiment to find out….Actually, we didn’t set out to run an experiment. We just wanted to create something fun that we thought our awesome tasking community would enjoy. The idea? Give people the chance to rate puppies’ cuteness in their spare time…There was a clear gender gap — a very consistent pattern of women rating the puppies as cuter than the men did. The gap between women’s and men’s ratings was more narrow for the “less-cute” (ouch!) dogs, and wider for the cuter ones. Fascinating.

I won’t even try to unpack the societal implications of these findings, but the lesson here is this: If you’re training an artificial intelligence model — especially one that you want to be able to perform subjective tasks — there are three areas in which you must evaluate and consider demographics and diversity:

  • yourself
  • your data
  • your annotators

This was a simple example: binary gender differences explaining one subjective numeric measure of an image. Yet it was unexpected and significant. As our industry deploys incredibly complex models that are pushing to the limit chip sets, algorithms and scientists, we risk reinforcing subtle biases, powerfully and at a previously unimaginable scale. Even more pernicious, many AIs reinforce their own learning, so we need to carefully consider “supervised” (aka human) re-training over time.

Artificial intelligence promises to change all of our lives — and it already subtly guides the way we shop, date, navigate, invest and more. But to make sure that it does so for the better, all of us practitioners need to go out of our way to be inclusive. We need to remain keenly aware of what makes us all, well… human. Especially the subtle, hidden stuff….(More)”

Doctors’ Individual Opioid Prescription ‘Report Cards’ Show Impact

Curated on Posted on by Stefaan Verhulst

Scott Calvert at the Wall Street Journal: “Several states, including Arizona, Kentucky and Ohio, are using their state prescription monitoring databases to send doctors individualized “report cards” that show how their prescribing of addictive opioids and other drugs compares with their peers.

“Arizona probably has the most complete one out there right now—it’s pretty impressive,” said Patrick Knue, director of the Prescription Drug Monitoring Program Training and Technical Assistance Center at Brandeis University, which helps states improve their databases.

Arizona’s quarterly reports rate a doctor’s prescribing of oxycodone and certain other drugs as normal, high, severe or extreme compared with the state’s other doctors in his medical specialty.

During a two-year pilot program, the number of opiate prescriptions fell 10% in five counties while rising in other counties, said Dean Wright, former head of the state’s prescription-monitoring program. The report cards also contributed to a 4% drop in overdose deaths in the pilot counties, he said.

The state now issues the report cards statewide and in June sent notices to more than 13,000 doctors statewide. Mr. Wright said the message is clear: “Stop and think about what you’re prescribing and the impact it can have.”
The report cards list statistics such as how many of a doctor’s patients received controlled substances from five or more doctors. Elizabeth Dodge, Mr. Wright’s successor, said some doctors ask for the patients’ names—information they might have gleaned from the database….(More)”