Data Science newsletter – March 12, 2021

[Rita Orij] said global leaders had the best interests of their people at heart, but the policies and precautionary measures put in place had no modern precedent.

“What we don’t understand is how are they actually working with people? Is there any way they are affecting people negatively or positively, especially when you think about emotional and mental health.

“So we gathered millions of tweets that people are commenting with the #stayathome hashtag, and we employed machine learning approaches to try to analyze these tweets and divide them into different emotions that are embedded in each of the tweets, using a different emotion classes.”

Education tech company Coursera filed its IPO prospectus with the Securities and Exchange Commission on Friday, and plans to list on the New York Stock Exchange under the symbol “COUR.”

The Mountain View, California-based company offers individuals access to online courses and degrees from top universities, a business that boomed during the Covid-19 pandemic.

When women and girls have the resources and opportunities to turn their potential into power, it changes the trajectory of their lives and strengthens entire communities. I’ve seen this play out first hand while living in India, where public health programs that put resources and decision-making in the hands of women drove much stronger outcomes for their families and villages. I’ve seen this in my own life, when bosses — both male and female — gave me stretch opportunities and bet on my leadership.

This is why I was excited to join our CEO Sundar Pichai to launch our global Google.org Impact Challenge for Women and Girls at a Google for India Women Will event earlier this morning. We’re calling on ideas from nonprofits and social organizations around the world that are working to advance the economic empowerment of women and girls and create pathways to prosperity. Google.org will provide $25 million in overall funding and Impact Challenge grantees will receive mentoring from Googlers, Ad Grants and additional support to bring their ideas to life.

The Association for Computing Machinery nixed Google as a sponsor for the fourth annual Conference on Fairness, Accountability and Transparency, happening this week online. The association had been facing pressure to address recent events at Google — namely the departure of two female computer scientists who called out bias in artificial intelligence and within the company.

Conference sponsorship co-chair Michael Ekstrand, assistant professor of computer science at Boise State University, said via email Monday that the event’s Executive Committee “concluded it was in the best interests of the community to pause the sponsorship relationship with Google while we revisit the sponsorship policy.” This means that Google is not included as a sponsor for the conference, he continued, “but has no other impact — researchers employed by Google are welcome to attend and present their work at the conference.”

In exclusive interviews with ESSENCE, Goldman Sachs shared plans for its One Million Black Women (OMBW) initiative, a $10 billion investment in support of Black women over the next 10 years. The initiative is named for and guided by the organization’s goal of impacting the lives of at least one million Black women by 2030.

Why? Because when Black women win, everybody wins.

For Goldman Sachs’ CEO David Solomon, investing in women-led businesses has been a values-driven focus since 2008, when the financial giant launched its 10,000 Women global program. Today, he’s taking that responsibility several steps forward. Solomon explains the foundation of One Million Black Women: “Given all that’s happened over the course of the last year, we’ve done a lot of talking at the firm—and even more listening—to help us figure out how we can do more to end the racial inequity and the gaps that have existed in society for well too long.”

It’s not exactly breaking news that far-right misinformation — better known to most as “lies” — tends to do well on Facebook. But it’s telling that the biggest takeaway from a new study that attempts to understand the phenomenon is that Facebook itself is our chief obstacle to understanding more.

New York University’s Cybersecurity for Democracy team released a paper on Wednesday bearing the title “Far-right sources on Facebook [are] more engaging.” The data isn’t terribly surprising if you’ve been paying any attention to the news of the past half-decade (and longer) and the role social media has played.

The report notes that content flowing out from sources rated by independent news rating services as far-right “consistently received the highest engagement per follower of any partisan group.”

In December 2020, software developer Angie Hinrichs at the University of California, Santa Cruz (UCSC), applied for access to a labor-saving data feed from GISAID, a nonprofit database of viral sequences including those of the pandemic coronavirus, SARS-CoV-2. She wanted GISAID’s data so she could display mutations on UCSC’s coronavirus Genome Browser. That tool ties any position in the virus’ nearly 30,000-letter genome to other scientific information, much as Google Maps shows gas stations and restaurants near addresses.

With more than 700,000 genomes from more than 160 countries, GISAID is by far the world’s largest database of SARS-CoV-2 sequences. Access to the free, nonprofit repository has become vital to Hinrichs and thousands of other scientists and public health agencies tracking the virus’ alarmingly rapid evolution.

But instead of getting a direct data feed, Hinrichs lost her existing access to two conveniently packaged GISAID files that are the next best thing.

As I write this, the world is probably days away from the “Great Email Robbery,” where a large number of threat actors around the globe are going to pillage and ransom the email servers of tens of thousands of businesses and local governments. Or at least pillage those that the purported Chinese actors haven’t already pillaged.

On Mar. 5, the investigative journalist Brian Krebs reported that an “unusually aggressive Chinese cyber espionage unit” had gained access to more than 30,000 U.S. organizations. The New York Times detailed on Mar. 6 that “The number of victims is estimated to be in the tens of thousands and could rise.” How did the attackers breach the companies? The Chinese actors developed a way to hack Microsoft Exchange and then attacked the organizations from there. And many of those attacked are still vulnerable to follow-on attacks not just by the Chinese but numerous criminals. The impact of the Exchange hack will certainly be greater than SolarWinds and researchers aren’t even close to the end of the story. But it’s a complicated story, with a lot to untangle.

The CEOs of Facebook (FB), Google (GOOG, GOOGL), and Twitter (TWTR) will appear later this month before a U.S. House subcommittee to face questions over the spread of misinformation tied to the 2020 election and COVID-19.

In a new interview, author Walter Isaacson — best known for his biography of late Apple (AAPL) CEO Steve Jobs — said social media platforms should take more responsibility for the extremism and misleading information fostered by their sites. He offered a blistering criticism of the algorithms that determine what users see, calling them “dangerous.”

The social media algorithms “spin us out of control” and “push us to be more and more extreme in our behavior,” says Isaacson, author of a new book entitled “The Code Breaker: Jennifer Doudna, Gene Editing, and the Future of the Human Race.”

Artificial intelligence will play a pivotal role in the future of health care, medical experts say, but so far, the industry has been unable to fully leverage this tool. A Yale study has illuminated the limitations of these analytics when applied to traditional medical databases — suggesting that the key to unlocking their value may be in the way datasets are prepared.

Machine learning techniques are well-suited for processing complex, high-dimensional data or identifying nonlinear patterns, which provide researchers and clinicians with a framework to generate new insights. Achieving the potential of artificial intelligence will require improving the data quality of electronic health records (EHR).

“Our study found that advanced methods that have revolutionized predictions outside healthcare did not meaningfully improve prediction of mortality in a large national registry. These registries that rely on manually abstracted data within a restricted number of fields may, therefore, not be capturing many patient features that have implications for their outcomes,” said Rohan Khera, MD, MS, the first author of the new study published in JAMA Cardiology. “We believe that the next frontier for improving clinical prediction may be the application of these methods to the high-dimensional granular data collected in the EHR.”

Exclusive: YouGov poll reveals extent of abuse and lack of faith in authorities’ ability to deal with it

Larger and more selective universities are going to like what they’re seeing in the latest college application data, summarized in a March 9 letter from Common Application to its member institutions. The Common App is a standardized form accepted by 900 colleges and universities, including all the Ivy League schools. Like other early indicators about where fall semester enrollments may end up, these figures suggest it could be a good year for institutions that have larger undergraduate enrollments (more than 10,000) and employ a selective admissions policy (admitting fewer than 50% of applicants).

The letter summarizes application data through March 1. To measure the differences between last year and this, comparisons are limited to colleges that were Common App members both years.

From speeding up diagnostic testing to streamlining communications between health-care facilities and promoting hand-washing, University of Toronto startups are finding a number of ways to contribute to the fight against COVID-19.

Many U of T entrepreneurs pivoted quickly during the early days of the pandemic to help address a global health threat, demonstrating a capacity for innovation, flexibility and quick-thinking.

With Entrepreneurship Week underway, here’s a look at six U of T startups whose innovative products and services are helping shape the response to the pandemic.

At the end of 2020, our behavioral science think tank, BEworks, completed one of the largest North American surveys of the public’s attitudes toward COVID-19 vaccination to date, covering over 3,700 nationally representative Canadians. Like other surveys in this sphere, we found that up to a third of respondents are opposed to getting the vaccine. But what really shocked us was how often individuals endorsed the belief that vaccination is unnecessary because the body’s “natural defenses” would do a better job at protecting it from infections.

These data, which have been condensed into our recently published report, have given us valuable insights into the psychological barriers to vaccine uptake, and prompted us to ask questions about the greater societal threat posed by pro-nature thinking. As it turns out, the appeal to nature fallacy has been resurfacing throughout recent history at great cost to humanity. And dissecting this story, as well as the fallacy’s role in the near future, will be crucial for our progress far beyond ending the COVID-19 pandemic.

Our findings may not be entirely shocking when we consider the myriad ways in which the appeal to nature fallacy has played out over the course of the pandemic.

Online March 23, starting at 9 a.m. Eastern. “18 speakers will explore applications of Machine Learning from both the business and technical areas of expertise.” [$$]

Online March 16-17. “Join your tech colleagues from higher education, industry, nonprofits, and government to learn about the ways UMass researchers are applying artificial intelligence, machine learning, and computing at scale to serve the common good, now and in the future.” [registration required]

Rejection emails made simple.

Get well-written AI-generated rejection emails that reduce your writing time by 95% so you can focus on the things you love.

I’m still curious about the NYT. It paints a certain picture of the world—and (in my circles) this ends up being the default picture, whether or not you agree with it.

I wanted to know more about this picture. So for the next few weeks I’ll be publishing a series of posts on the New York Times, drawing on data scraped from their front page and pulled from their official API.

This first post is all about A/B testing: how the NYT tests different headlines and how they change over time.