NYU Data Science newsletter – November 6, 2015

GROUP CURATION: N/A

Machine Learning APIs make it easy for developers to develop predictive applications. Here we review 5 important Machine Learning APIs: IBM Watson, Microsoft Azure Machine Learning, Google Prediction API, Amazon Machine Learning API, and BigML.

This channel brings together all of the F1000Research articles describing new R packages that are applicable to the life sciences. R is the statistical language + environment of choice for many researchers due to the fact that it is open source, platform-independent and allows any user to extend its functionality. Bundling this new code together with documentation and test data into an R package makes it much easier to download, use and, importantly, reproduce.

Volkswagen’s infamous emissions-test-subverting software lurked in cars for years before it was discovered by regulators. The company got away with it for so long, in part, because it’s hard to actually tell what’s going on within the embedded computers of an automobile.

One way to deal with the issue would be to require that certain types of companies, like automakers, release the software code that powers their products to the public, so that researchers could evaluate deceitful practices as well as security flaws. A less extreme solution, suggested by Zeynep Tufekci in the New York Times this year, would be to simply require automakers to release code to auditors, the same way the manufacturers of casino slot machines must open their code to gambling regulators.

SFI Board of Trustees Chair Michael Mauboussin interviews Daniel Kahneman, winner of the 2002 Nobel Prize in Economics

Privacy is an increasingly important social implication of technology, and we spend quite a bit of time about it in our required undergraduate ethics and social implications of technology class, CS 4001. Since we’re talking about privacy, it makes sense to talk about surveillance. Since 2004, I’ve taught a class about The USA PATRIOT Act, and more recently I’ve added a class on information revealed by Edward Snowden. I spend more time preparing for those classes than for any other two or three put together—it’s confusing and complicated. There are provisions of the Patriot Act that are absolutely essential—like broadening the jurisdiction of warrants to tap phones to the entire country (rather than making you get a warrant in each state). And others that are egregious violations of our liberty—like the section 215 provision that lets the government get the records of any organization without a warrant or probable cause and bars the organization from acknowledging the search. The FBI can simply demand the membership list of a mosque—and they have done so. For the last two years, I’ve assigned my students to watch the PBS Frontline documentary United States of Secrets, about US warrantless surveillance (“The Program”) and information leaked by Edward Snowden. In our class discussion, we don’t focus on Snowden, but on other people—like NSA analyst Thomas Drake—and the tough decisions they had to make. After class on Tuesday where I carefully spell out what’s allowed under the Patriot Act and the Foreign Intelligence Surveillance Act (FISA), I feel like a bit of fool on Thursday when we discuss The Program and the fact that all those rules aren’t really followed anyway.

I do my best not to express any opinions to my class—I present the facts, and ask them what they think. And as much as possible, I emphasize tradeoffs and try to show the issues as complicated. And then I walk back from class and scratch my head—what do I actually think?

A recent report by Frost and Sullivan discussed how big data and analytics are helping to power the retail industry. It is creating opportunities for growth and stability in this $22 trillion industry. Not only in the e-commerce space where complex data has been at the core of its success, but also in improving the fortunes of brick and mortar stores.

Data has played a key role in many brick and mortar stores, but today it is going beyond what has previously been considered. For instance, the interlinking of data sources has given retailers new opportunities to identify trends and more importantly, the root causes of them. This could be anything from the way that a display looks as customers walk past the window, through to how they walk through the store once they are inside.

Jeremy Bailenson, director of the Stanford Virtual Human Interaction Lab and a professor in Stanford University Department of Communication, discusses the ways in which virtual reality is affecting communication, learning and the tools society can leverage to address a range of global issues.

Midlife increases in suicides and drug poisonings have been previously noted. However, that these upward trends were persistent and large enough to drive up all-cause midlife mortality has, to our knowledge, been overlooked. If the white mortality rate for ages 45?54 had held at their 1998 value, 96,000 deaths would have been avoided from 1999–2013, 7,000 in 2013 alone. If it had continued to decline at its previous (1979?1998) rate, half a million deaths would have been avoided in the period 1999?2013, comparable to lives lost in the US AIDS epidemic through mid-2015. Concurrent declines in self-reported health, mental health, and ability to work, increased reports of pain, and deteriorating measures of liver function all point to increasing midlife distress.

Code fraud is the implementation of software that induces illegal, unethical, or otherwise malicious outcomes. Difficult to detect, code fraud is a growing concern across industries. Detecting it is a challenge that vastly outpaces our current regulatory apparatus. In the case of Volkswagen, a small non-profit (the International Council on Clean Transportation), not regulators or internal whistleblowers, pieced the complex chain of dependencies together and came forward. Earlier this year, a report written by Forensiq, a small firm, gave notice to Google that apps available through their app store contained malicious code designed to defraud advertisers. Eric Scott Hunsader, a developer for the Nanex exchange is still wondering why regulators have not forced the New York Stock Exchange to investigate—and close—specific loopholes being exploited by high-frequency trading firms. It could be, as he notes, because “Code fraud is *integral* to how high frequency trading works.”

The code fraud committed by Volkswagen is unique not because it went on for years, not because it outwitted regulators, not because it generated collateral damage for unwitting bystanders, not because it is worth hundreds of millions of dollars, but because it was detected at all.