Welcome to the December 20, 2019 edition of ACM TechNews, providing timely information for IT professionals three times a week.
ACM TechNews mobile apps are available for Android phones and tablets (click here) and for iPhones (click here) and iPads (click here).
To view "Headlines At A Glance," hit the link labeled "Click here to view this online" found at the top of the page in the html version.
The online version now has a button at the top labeled "Show Headlines."
|
|
|
Trump Picks Computer Scientist to Lead NSF
Nature
Alexandra Witze; Lauren Morello
December 19, 2019
U.S. President Donald Trump has nominated computer scientist and Arizona State University vice president Sethuraman Panchanathan to be director of the U.S. National Science Foundation (NSF). Kelvin Droegemeier, head of the White House Office of Science and Technology Policy, cited the nominee's dedication, creativity, and insight as critical for maintaining NSF's leadership in discovery and exploration. Panchanathan is on the National Science Board, and his research has concentrated on the interface between people and computers—including development of assistive technologies like haptic computing. If the nomination is approved, Panchanathan would succeed France Córdova, whose six-year term expires next year.
|
Facebook Wins Landmark Case as EU Finds Data Transfers Legal
The Telegraph (U.K.)
Hasan Chowdhury
December 19, 2019
An adviser to the European Court of Justice (ECJ) declared the transfer of data outside the European Union (EU) to be legal. ECJ advocate general Henrik Saugmandsgaard Oe's non-binding opinion refutes Austrian lawyer Max Schrems' claim that Facebook's data-transfer protocol via contractual clauses breaches EU users' privacy because it lacks sufficient online protection. Schrems said he was "generally happy" with the results of the case. “Everyone will still be able to have all necessary data flows with the U.S., like sending emails or booking a hotel in the US,” he said, adding. “It is really upon the United States to ensure baseline privacy protections for foreigners. Otherwise, no one will trust U.S. companies with their data.”
|
Model Suggests Hidden Personalities of Jobs
University of Melbourne
Lito Vilisoni Wilson
December 17, 2019
A study by researchers at Australia's University of Melbourne, University of Technology Sydney (UTS), and University of New South Wales suggested defining the personality traits and values of different occupations could be critical to matching people with their ideal professions. The researchers examined more than 128,000 Twitter users representing some 3,500 occupations; a combination of artificial intelligence, machine learning, and data analytics yielded a data-driven vocation compass to recommend careers for certain personalities. UTS' Marian-Andrei Rizoui said the model could recommend an occupation that lined up with personality traits with more than 70% accuracy. Said Rizoui, “Even when the system was wrong it was not too far off, pointing to professions with very similar skill sets.”
|
Voting Machine Parts Made by Foreign Suppliers Stir Security Concerns
The Wall Street Journal
Alexa Corse
December 16, 2019
A report from supply-chain monitoring consultancy Interos found that voting-system vendors could be at risk of using insecure parts manufactured by companies associated with China and Russia. Interos used one voting machine it analyzed as a case study, finding about 20% of its parts—including processors, software, and touchscreens—were sourced from China-based companies. Vendors of voting machines assailed the research; in response, Interos CEO Jennifer Bisceglie said, "All the report was trying to do was trying to elevate the conversation around the fact that every company and every country is hyperconnected.”
*May Require Paid Registration
|
IT Can Save Police Lives: Study
University of Houston News
Jeannie Kever
December 11, 2019
A study by University of Houston (UH) and Temple University researchers found U.S. law enforcement agencies can significantly reduce the number of police officers killed or injured in the line of duty through the use of information technology (IT). UH's Paul A. Pavlou and Temple's Min-Seok Pang assembled data from the U.S. Federal Bureau of Investigation, Bureau of Justice Statistics, and the U.S. Census to build a dataset that correlated IT use and reported violence against law enforcement from 4,325 police departments over six years. The research found that extensive IT use by police could lower violence against officers by 42% to 50%. The researchers said the use of IT to learn more about potential suspects improves the probability that police can make an arrest without being subjected to violence.
|
Apple, Google, Amazon Decide to 'Play Nice' Over Smart Home Tech
BBC News
December 18, 2019
Apple, Google, and Amazon have announced a partnership to improve smart-home technology's ease of use by creating a new standard to ensure their smart products are compatible with all three companies’ smartphones and voice assistants. The three tech giants also will work with the Zigbee Alliance, and a successful effort could remove consumers and manufacturers' burden of favoring one smart-home technology over others. Draft specifications for the new standard are not expected before late 2020. The companies said current smart-home products should continue to operate after the new standard is implemented. The first product category to be targeted by the new standard will be smart home security products.
|
Smart Intersections Could Reduce Autonomous Car Congestion
Cornell Chronicle (NY)
Melanie Lefkowitz
December 16, 2019
Cornell University researchers developed a smart intersection control model to boost vehicle capacity on urban streets, cut congestion, and minimize accidents with autonomous cars. The model permits platoons of autonomous cars to pass through one-way intersections without waiting; a microsimulation found the model facilitated an increase of vehicle capacity on city streets of as much as 138% versus conventional traffic signal systems. The model assumes only self-driving cars are on the road, and the researchers are considering scenarios with both autonomous and human-driven cars for future research.
|
Many Facial Recognition Systems Are Biased, Says U.S. Study
The New York Times
Natasha Singer; Cade Metz
December 19, 2019
A study by U.S. National Institute of Standards and Technology (NIST) researchers found most commercial facial recognition algorithms are biased, and falsely identify African-American and Asian faces 10 to 100 times more frequently than Caucasian faces. The error rates were highest in identifying Native Americans, while identifying women was even more problematic than for men. The researchers tested 189 algorithms from 99 developers on a law enforcement dataset containing over 18 million photos of 8.5 million people. The facial-matching algorithms used in law enforcement had the highest error rates for African-American women. NIST scientist Patrick Grother said he hopes these findings will encourage developers to consider ways to recognize such bias and work to remedy it. Carnegie Mellon University's Maria De-Arteaga said, "We have to think about whether we really want these technologies in our society."
|
Sensing Technology Could Improve Machine Learning Precision for Manufacturing, Electric Vehicles, Smart Homes
Purdue University News
Chris Adam
December 11, 2019
Purdue University researchers have developed a sensor that tracks electrical current, which could be used with machine learning to track currents in applications ranging from electric cars to manufacturing and home design. The device can transmit the current data it captures to any computing system via Bluetooth, USB, or other techniques. The researchers used a machine learning algorithm with the sensor to help it gather and interpret energy data. Purdue's Byunghoo Jung said, "This sensor could be used with machine learning to train manufacturing robots, provide precise tips for homeowners on cutting down their energy usage, or help diagnose issues with electric vehicles and scooters."
|
U.K. to Create Regulator to Police Big Tech Companies
Financial Times
Madhumita Murgia; Kate Beioley
December 19, 2019
The U.K. government in 2020 will establish a regulator to police major technology companies following Brexit. The new regulator will have the ability to implement new rules for technology firms, including an enforceable code of conduct and expanded consumer data accessibility. Antony Walker of non-profit TechUK said, “Competitive markets are extremely important to ensuring consumer choice and driving up standards. TechUK believes an informed regulator is vital to securing this.” Antitrust experts think the regulator is necessary, as are reforms to digital competition laws. Critics are concerned about the scope of any such potential intervention.
*May Require Paid Registration
|
Tool Uses AI to Flag Fake News for Media Fact-Checkers
Waterloo News
December 16, 2019
Researchers at the University of Waterloo in Canada have developed an artificial intelligence tool that could help social media networks and news organizations screen out fake news. The tool employs deep learning algorithms to ascertain whether assertions made in posts or stories are backed by other posts and stories on the same topic, with 90% accuracy. Waterloo's Alexander Wong said the tool can augment the capabilities of human fact-checkers by flagging information “that doesn’t look quite right for verification. It isn’t designed to replace people, but to help them fact-check faster and more reliably.” Said Waterloo’s Chris Dulhanty, “We need to empower journalists to uncover truth and keep us informed. This represents one effort in a larger body of work to mitigate the spread of disinformation.”
|
Password Meters Could Increase Risk of Cyberattack
University of Plymouth (UK)
Alan Williams
December 19, 2019
A study by the University of Plymouth in the U.K. evaluating 16 popular password meters found these tools offer "inconsistent and misleading" advice that could elevate the risk of cyberattack. The study by Plymouth's Steve Furnell concentrated on dedicated password meter websites, but also assessed meters embedded in online services like Dropbox and Reddit, as well as those included as standard features on certain devices. The analysis revealed varying advice; while some meters direct users toward more secure account passwords, others do not flag extremely insecure passwords. Said Furnell, "Credible password meters can have a valuable role to play, but misleading meters work against the interest of security and can simply give further advantage to attackers."
|
|
