Data Science newsletter – July 6, 2018

GROUP CURATION: N/A

Berners-Lee has seen his creation debased by everything from fake news to mass surveillance. But he’s got a plan to fix it.

To fully understand biodiversity and how it is changing, you need to look near, far, and in-between, according to a new study.

Researchers at Yale University studied 50 years of data about nesting birds in North America and tracked biodiversity changes on a local, regional, and continental scale. They found significant differences in how much change had occurred, based upon how wide a geographic net they cast.

The findings have implications for how to assess biodiversity in a rapidly changing world, as well as how biodiversity information should be presented. The study appears July 2 in the journal Nature Communications.

China represents both a threat and an opportunity for U.S. technology companies. This week, we saw Google invest $550 million in China’s e-commerce provider JD.com while Qualcomm introduced new modems and reference designs for kids’ smartwatches based primarily on interest from the Chinese market.

So I read with interest this week the report from the GSMA on industrial IoT in China. Most of the report is data collected by a variety of consulting firms, but there are worthwhile stories from big Chinese mobile operators. It is the GSMA, after all.

One of the biggest themes in the report is the use of cellular NB-IoT, which Chinese operators are deploying at a rapid pace.

In a decision that drew scant attention last fall, each of Pennsylvania’s 14 state-owned universities gained authority to increase dramatically the tuition and fee revenue they can plow back into scholarships, up to $50 million more collectively each year.

Given that the base tuition at those State System of Higher Education campuses is $7,492 a year, it does not take a math whiz to see the potential impact of that Nov. 13 vote by the system’s board of governors.

Now that State System administrators have proposed scrapping the single tuition rate across the system, instead letting each of the 14 universities set their own prices and apply financial aid as they see fit, exactly how schools use that new scholarship authority looms even larger.

Google is turning startup investor to further its goal of putting Google services like search, maps, and its voice assistant front and center for the next billion internet users in emerging markets. It has invested $22 million into KaiOS, the company that has built an eponymous operating system for feature phones that packs a range of native apps and other smartphone-like services. As part of the investment, KaiOS will be working on integrating Google services like search, maps, YouTube and its voice assistant into more KaiOS devices, after initially announcing Google apps for KaiOS-powered Nokia phones earlier this year.

“This funding will help us fast-track development and global deployment of KaiOS-enabled smart feature phones, allowing us to connect the vast population that still cannot access the internet, especially in emerging markets,” said KaiOS CEO Sebastien Codeville in a statement.

Facebook has acquired the team behind Bloomsbury AI, a London-based start-up, the company announced on Tuesday.

Founded in 2015, Bloomsbury AI specialises in natural language processing technology, and has developed an AI called “Cape,” which can read documents and then answer questions about their contents.

Despite a number of attempts to apply it to tackle poverty, famine or displacement, “we have yet to see successful stories of machine learning truly advancing development”, according to Maria De-Arteaga, of the Machine Learning and Public Policy program at Carnegie Mellon University in Pittsburgh Pennsylvania in the United States.

De-Artega spoke at the UNESCO conference Tech4Dev in Switzerland last week (27-29 June), where other researchers presented some projects that show its promise.
Stefano Ermon, Assistant Professor in the Department of Computer Science at Stanford University in the US state of California, told the meeting that he believes machine learning could solve the “data drought” in developing countries.

“We are getting access to a lot of new data streams from satellites, from social media, from phones,” he said. “They clearly contain a lot of information about the outcomes we care about … they have largely not been used because they are massive and they are unstructured.

An international team of astronomers has discovered an ancient and dramatic head-on collision between the Milky Way and a smaller object, dubbed the “Sausage” galaxy. The cosmic crash was a defining event in the early history of the Milky Way and reshaped the structure of our galaxy, fashioning both its inner bulge and its outer halo, the astronomers report in a series of new papers.

The astronomers propose that around 8 billion to 10 billion years ago, an unknown dwarf galaxy smashed into our own Milky Way. The dwarf did not survive the impact: It quickly fell apart, and the wreckage is now all around us.

“The collision ripped the dwarf to shreds, leaving its stars moving in very radial orbits” that are long and narrow like needles, said Vasily Belokurov of the University of Cambridge and the Center for Computational Astrophysics at the Flatiron Institute in New York City. The stars’ paths take them “very close to the center of our galaxy. This is a telltale sign that the dwarf galaxy came in on a really eccentric orbit and its fate was sealed.”

The National Science Foundation (NSF) has awarded Howard University’s College of Engineering and Architecture (CEA) a three-year, $1,000,000 grant to fund an innovative cyber security research project created by Associate Dean Moses Garuba, Ph.D. and Associate Professor of Electrical Engineering and Computer Science Danda B. Rawat, Ph.D.

The research project, entitled, “Security Engineering for Resilient Mobile Cyber-Physical Systems,” will focus on significantly advancing the field of cybersecurity for networked systems.According to the NSF award abstract, the goal is to design, develop and evaluate the cyber-defense solutions for resilient cyber-physical systems using a federated framework.

A boom in artificial intelligence research has drawn the tech industry’s biggest companies and their checkbooks to the storied English city.

The internet was meant to make the world a less centralised place, but the opposite has happened. Ludwig Siegele explains why it matters, and what can be done about it

Field data collection is challenging in the best of conditions, and in an environment as harsh as Antarctica, large-scale, long-term field monitoring studies are rare.

To get around the problem, a multinational collaborative research effort has installed time-lapse cameras at more than 30 penguin breeding colonies in Antarctica and remote islands in the Southern Ocean. The network of cameras takes images of the penguins year-round, enabling researchers to monitor the health of the colonies over time and space by documenting nest survival rates and comparing changes in population dynamics and reproduction to shifts in temperature and human fishing activity. All without people being present.

The planned Global Cybersecurity Institute at Rochester Institute of Technology got a tremendous boost with the receipt of $5 million grant through the New York State Higher Education Capital Matching Grant Program.

The grant, combined with designated funding from a donation made by 2009 RIT alumnus Austin McChord, will go toward construction of a 45,000-square-foot building to house the new institute. The institute will address the critical workforce needs in cybersecurity through education and professional development programs and will conduct research to advance the fields of cybersecurity and artificial intelligence.

Last year, when I decided to survey college graduates who had majored in English, I was looking for help in guiding my own students on life and work after college. One of the anonymous graduates who replied was “Casey,” who made clear that — at age 25, and two years out of college — she had hit a “dead end” and considered majoring in English the “Worst Decision of My Life.”

Having excelled as a student, she now works at a coffee shop, owes $40,000 in student loans, relies on government aid for food and health care, and sees no prospect of getting a better-paying job or borrowing more for grad school. To Casey, the folks buying the coffee seem to “have more money and hope than an English major will ever have.” She advised all English majors to switch to a STEM field.

Casey is not alone in her regret, according to a vast recent study by Gallup and Strada, “On Second Thought: U.S. Adults Reflect on Their Education Decisions.” Talking to a representative, random sample of 89,492 adults in the United States, researchers found that nearly half (48 percent) of B.A.s in the liberal arts would, if they had it to do over again, pursue a different area of study.

A central truism of U.S. privacy law is that if you share information, you do not have an expectation of privacy in it. This reasoning runs through both Fourth Amendment jurisprudence and privacy tort cases, and has repeatedly been identified as a central failing of American privacy law in the digital age.1 On June 22, in Carpenter v. United States,2 the Supreme Court did away with this default. While repeatedly claiming to be fact-bound and incremental,3 Chief Justice Roberts’s opinion has paradigm-shifting implications not only for Fourth Amendment law, but also for private-sector privacy law.

In short, the Court in Carpenter has declared that Big Data is different. Just how different remains to be seen.

Bethesda, MD July 23. “The workshop will feature leading experts from Amazon and IBM, and scientists from the NIH Clinical Center, Massachusetts Institute of Technology, Stanford University, and the University of Cincinnati who are employing AI/ML in biomedical research settings.” [sold out, waitlist available]

Cambridge, MA July 11, starting at 7 p.m., Stata Center (MIT). “We invite you to join a conversation on the new tech resistance. Our panel will explore where this moment of mobilization came from and how we can help it grow.” [free]

Vancouver, BC, Canada August 3 at 515 West Hastings Street. [$$]

Here’s the good news: just because we can’t always A/B test a major experience doesn’t mean we have to fly blind when it matters most. A range of econometric methods can illuminate the causal relationships at play, providing actionable insights for the path forward.

Kaggle’s public data on competitions, users, submission scores, and kernels

While the furor around Facebook’s Cambridge Analytica scandal has died down, the world’s largest social network continues to make big changes to its APIs.

Today, the company announced that its PagesTrack this API, MarketingTrack this API, Lead As Retrieval, and Live VideoTrack this API APIs will only be accessible to apps that have been reviewed. Such reviews are part of the efforts Facebook is making “to better protect people’s information” by reducing the risk that apps will collect and use data in ways that they are not authorized to.

In some cases, app review oversight is being added to APIs that had temporarily been shuttered.