Data Science newsletter – May 4, 2021

Scientists have few categories at their disposal to describe droughts, which are more complex than mere shortages of precipitation or surface water. For example, some local shortages can be invisible, as when water is transferred into a dry area from a distant source. Other shortages are chronic, with communities continuously requiring more water than is available, even in wet years. Some water shortages occur when water quality becomes so degraded that even though there may be plenty of water available, little of it is usable. With such variation in conditions, scientists need better language to conceptualize droughts.

To address this deficiency, AghaKouchak et al. propose the idea of anthropogenic drought, which accounts for both natural variation and human actions. For instance, droughts are affected by local decisions about water and land use as well as by such global conditions as greenhouse gas levels and climate change. The authors suggest that scientists think of drought as a process with contributing factors, effects, and feedbacks rather than as simply a final product.

De-recommendation does seem good at suppressing a subset of misinforming/harmful content, but we don’t know the full set of treated content. Likewise, we have some concern about the ideological distribution of boosted mainstream media. E.g., is Fox News amplified? 2/5

NFL coaches love to talk about the concept of complementary football. Offense and defense feeding off each other’s success toward the common goal of victory.

As it turns out, complementary mathematics can be just as crucial on the analytics side, as two University of Pittsburgh doctoral students recently proved in the league’s 2021 Big Data Bowl, a massive data science competition in which more than 200 teams from all over the world compete to impress a panel of judges — including coaches — with fresh ideas.

Wei Peng and Marc Richards did just that in collaborating on the winning entry, crowned last month during a slickly produced Zoom broadcast on which they presented their models measuring the effectiveness of defensive backs.

A so-called smart petri dish prototype is a collaborative project’s first proof of concept for thermoforming plastic parts with printed sensor systems. The development is said to offer a pathway for transforming plastic parts through the use of printed electronics.

The collaboration between Greiner Assistec, part of Austria-based Greiner Packaging, and accensors, a startup business unit of innoME GmbH based in Munich, aims to achieve the affordable, high-volume production of thermoformed plastic parts with printed sensor systems, said Greiner Assistec. “As part of the pilot project, an initial demo prototype has now been developed, taking the form of a smart petri dish,” explained the company in a press release. “This involves printing two sensors on the film before putting it through a thermoforming process. And the resulting solution also allows for ongoing monitoring, with the initial prototype making it possible to check the pH value and temperature of the medium in the dish using an accensors scanner and app. The resulting data obtained via this process can be harnessed to gain new insights in the realm of research and development,” added Greiner Assistec.

UCLA will establish a curriculum about social justice with a focus on data using a $5 million grant.

The curriculum will offer students a data-driven approach to researching, teaching and learning about social justice, said UCLA spokesperson Katherine Alvarado in an emailed statement.

The grant, donated by the Andrew W. Mellon Foundation, will fund the creation of a new undergraduate cluster course called “Data, Society, and Social Justice,” summer workshops, a master’s degree program in data and society and other courses and research opportunities, said Darnell Hunt, a sociology and African American studies professor and the dean of social sciences.

Admissions officers and enrollment managers are asking themselves several questions. Should they expect things to return to “normal” pre-COVID patterns this year or permanently alter their expectations? Should they change admissions or scholarship criteria? Should they throw out the predictive models they trained on past data after an unprecedented year? And if they keep existing processes and tools, how can they work with data scientists to recalibrate them to remain useful?

I believe predictive models still offer a lot of value to universities. For one thing, models trained on past data can be especially useful in understanding how reality differed from expectations. But the last year has revealed just how important it is that we fully understand the “how” and the “why” of the predictions these tools make about “who” is most likely to enroll or may need additional services to help them succeed at an institution.

The ongoing Covid-19 pandemic and its imposed distancing measures means touch between two people has become taboo. Shaking hands and hugging now seem like behaviour relegated to the past. A project carried out by the University of Applied Sciences and Arts of Southern Switzerland (SUPSI), in collaboration with the Hongik University in Seoul, is trying to imagine a “touch-free” society that maintains a sense of “real” contact while remaining as risk-free as possible. This is made possible via intelligent interfaces based on machine learning.

These interfaces aim not only to encourage compliance with protective measures, such as wearing a mask and keeping your distance, but also to completely reinvent public spaces.

Scientific data have traditionally been distributed via downloads from data server to local computer. This way of working suffers from limitations as scientific datasets grow toward the petabyte scale. A “cloud-native data repository,” as defined in this article, offers several advantages over traditional data repositories—performance, reliability, cost-effectiveness, collaboration, reproducibility, creativity, downstream impacts, and access and inclusion. These objectives motivate a set of best practices for cloud-native data repositories: analysis-ready data, cloud-optimized (ARCO) formats, and loose coupling with data-proximate computing. The Pangeo Project has developed a prototype implementation of these principles by using open-source scientific Python tools. By providing an ARCO data catalog together with on-demand, scalable distributed computing, Pangeo enables users to process big data at rates exceeding 10 GB/s. Several challenges must be resolved in order to realize cloud computing’s full potential for scientific research, such as organizing funding, training users, and enforcing data privacy requirements.

Neurolutions, a company with offices in Santa Cruz, CA and St. Louis, MO, won FDA de novo authorization to introduce its IpsiHand stroke rehab system in the United States. This is the first approval of a brain-computer interface technology for the rehabilitation of stroke patients, and it promises a faster and more complete recovery of hand function in many patients.

This week, UCS led a group of 16 scientific, professional, and academic organizations, including the American Public Health Association (APHA), the American Geophysical Union (AGU), Ciencia Puerto Rico, 500 Women Scientists, and the American Association of University Professors, that sent a letter to 24 federal science agencies and departments in support of diversifying the science advisory committees that support policy decisions across the government.

The letter follows from the Biden administration’s January 27 “Memorandum on Restoring Trust in Government Through Scientific Integrity and Evidence-Based Policymaking.” As President Biden strives to create the most diverse cabinet in history and seeks to incorporate diversity, equity, inclusion, and justice (DEIJ) into federal initiatives, it is just as important that external science advisors adequately reflect the makeup of our country in race, ethnicity, gender, sexual orientation and other identities and types of expertise. Having balanced representation will not only improve the evidence-based recommendations given to our decisionmakers but will mean that underrepresented voices and expertise will have a seat at the advisory table and the decisionmaking table.

“It doesn’t matter how poor, it doesn’t matter how wealthy, the racial disparities exist for all African-Americans and other people of color,” said Paul Mohai, a professor of environmental justice at the University of Michigan who was not involved with the study. He’s researched racial disparities in the distribution of hazardous waste dumps, industrial facilities and air pollution at schools — and why and how these disparities exist.

“When you look at the impact of air pollution in the U.S. from all the sources that contribute to fine particulate matter in the atmosphere, there is an overall systemic bias against people of color,” said study co-author Jason Hill, a biosystems engineering professor at the University of Minnesota.

A partnership between UC Davis and Maurice J. Gallagher, Jr., chairman and CEO of Allegiant Travel Company, has led to a new rapid COVID-19 test.

A recent study published Nature Scientific Reports shows the novel method to be 98.3% accurate for positive COVID-19 tests and 96% for negative tests.

“This test was made from the ground up,” said Nam Tran, lead author for the study and a professor of pathology in the UC Davis School of Medicine. “Nothing like this test ever existed. We were starting with a clean slate.” [video, 2:22]

Back in 2014, when researchers affiliated with Drexel University’s University Research Computing Facility (URCF) were deciding the name of its first high-performance computer cluster, they went with an ancient, mythological choice: Proteus, the shape-changing sea god from Greek mythology.

When naming its replacement computer cluster, the URCF went with something a little closer to home.

Picotte, which opened in February, is named in honor of Susan La Flesche Picotte (1865-1915), MD, who, according to current research, was the first Native American to become a physician in the U.S. Picotte graduated in 1889 from one of the Drexel University College of Medicine’s predecessor schools, Woman’s Medical College of Pennsylvania (WMCP), one of the world’s first medical schools for women and one that graduated many trailblazing doctors.

Chancellor Kevin Guskiewicz said Apple’s announcement aligns with the launch of UNC’s data science initiative. Few details have been released yet, but Jeffay said it includes a proposal for a school of data science.

“Carolina has created an environment where innovation can thrive, and we are proud of the foundation we have developed in our region for this type of investment,” Guskiewicz said in a statement. “This is a huge win for our region and the state of North Carolina and I’m looking forward to the opportunities this provides our students and faculty researchers for years to come.”

Light-sheet fluorescence microscopy, an imaging tool that can rapidly produce 3D images of complex cellular structures, gives scientists the power to visualize the myriad miniature dramas unfolding inside living cells and tissues. Scientists at Washington University School of Medicine in St. Louis are using the technique to watch, in astonishing detail, tiny organelles rearrange themselves inside cells, to monitor the stabilization of blood vessel walls in developing fish, and to map the network of filtration units in a human kidney. But high-resolution imaging generates reams of data, such that handling the enormous datasets has emerged as a chokepoint on the path to widespread adoption of the powerful imaging technique.

With the aid of a $1.2 million grant from the Arnold and Mabel Beckman Foundation, a team of researchers led by David W. Piston, PhD, the Edward Mallinckrodt Jr. Professor and head of the Department of Cell Biology & Physiology, is working to develop the computational tools necessary to analyze light-sheet microscopy data so scientists can take full advantage of its potential to help answer pressing biomedical questions.

We recently released RecSim NG, the “Next Generation” of simulators for recommender systems research and development. RecSim NG is a response to a set of use cases that have emerged as important challenges in the application of simulation to real-world problems. It addresses the gap between simulation and real-world applications, ensures the models are flexible and easily extendible, enables probabilistic inference of user dynamics, and addresses computational efficiency.

With more than 300 million users completing more than 7 billion exercises each month, language-learning platform Duolingo offers an example of data science methodology in action. Not only do the company’s enormous databases inform tweaks to Duolingo’s user experience and underlying infrastructure all the time, but the company’s data science teams conduct regular research into everything from optimizing reminder notifications to theories on how to improve teaching practices and outcomes for learners of indigenous languages.

Duolingo’s data science methodology underpins much of this work. To learn more about the nuts and bolts of how a project moves from an amorphous idea to a usable tool or valuable insight, Lead Data Scientist Erin Gustafson — one of RE•WORK’s Top 30 Women Aiding AI Advancement back in 2019 — took us through her team’s best practices.

Tracy Ring, managing director of cognitive, analytics and RPA alliances for Deloitte Consulting and founder of Deloitte’s women in data science and analytics group, spoke to InformationWeek about her team’s accomplishments and what organizations can do to improve gender parity especially in the data science workforce.

Ring says it has become part of her job to not only help organizations with their technology implementations but to also ensure diverse practitioners are involved.

Plot’s concise API and thoughtful defaults are designed for a more joyful visualization process, try out the beta and let us know!