Welcome to the November 11, 2020 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."
|
|
|
Software Focuses on Better Targeting Mental Health Resources
NC State News
November 9, 2020
Prototype software developed by researchers from North Carolina State University (NC State) and North Carolina Agricultural and Technical State University is designed to help policymakers and healthcare providers better address the mental health challenges of U.S. Hispanics. The decision-support system features two computational modules to help users determine the best ways to prepare for future mental health challenges, and identify the best courses of action to mitigate them. A predictive module employs multi-sourced state-level data to forecast the percentage of the Hispanic population that may suffer from anxiety and depression in the following week. The more flexible Markov Decision Process module helps users prescribe the most effective course of action for reducing the number of people experiencing those symptoms.
|
DNS Cache Poisoning Ready for Comeback
UC Riverside News
Holly Ober
November 11, 2020
Computer security researchers at the University of California, Riverside (UC Riverside) and China's Tsinghua University found critical security flaws that could lead to a resurgence of Domain Name System (DNS) cache poisoning attacks. The exploit derandomizes the source port and works on all cache layers in the DNS infrastructure, including forwarders and resolvers. The research team confirmed this finding by using a device that spoofs Internet Protocol (IP) addresses and a computer that can trigger a request out of a DNS forwarder or resolver; it exploited a novel network side channel to execute the attack. The team, which has demonstrated the exploit against popular public DNS servers, recommended the use of additional randomness and cryptographic solutions to combat it.
|
AI Shows Potential to Gauge Voter Sentiment
The Wall Street Journal
Jared Council; John McCormick
November 6, 2020
Some technology experts believe using artificial intelligence (AI) to gauge voter sentiment could help to better understand the electorate. Allen Institute for AI CEO Oren Etzioni said, "I would direct [pollsters] to try to leverage machine learning, data mining, and AI in their work more to get better projections." Heidi Messer with AI and predictive technology provider Collective[i] said polls must use data sources that capture actual behavior, rather than relying on historical classifications and averages. Italian software company Expert.ai used its natural language processing system to analyze millions of social posts around the U.S. presidential candidates, weighing factors like tone and emotion, and projecting how that might translate into votes; its predictions have closely matched actual results. Also yielding accurate forecasts is swarm-intelligence software from Unanimous.ai, which aggregates predictions and decisions from groups of people.
*May Require Paid Registration
|
Thin Holographic Video Display for Mobile Phones
IEEE Spectrum
Charles Q. Choi
November 10, 2020
At South Korea's Samsung Advanced Institute of Technology, researchers have invented a method for creating a thin holographic video display that may eventually enable 4K three-dimensional videos on mobile phones. The display has a special backlight with a beam deflector that can tilt the angles of coherent light beams from laser diodes, expanding the viewing angle 30-fold without increasing the number of pixels required. The Samsung researchers also used a slim geometric phase lens to gather scattered light from the pixels, shrinking the optical components' thickness to just 1 centimeter. The display’s single-chip holographic video processor can perform about 140 billion operations per second to generate 4K holographic color images at a speed of 30 frames per second.
|
Keep the Data Coming
KAUST Discovery (Saudi Arabia)
November 8, 2020
A preemptive memory management system developed by researchers at Saudi Arabia's King Abdullah University of Science and Technology (KAUST) can accelerate data-intensive simulations 2.5-fold by removing delays caused by slow data delivery. The KAUST team designed its multilayer buffer system (MLBS) to proactively maintain data as close as possible to the computing hardware by coordinating data flow among memory layers. KAUST's Tariq Alturkestani said, "MLBS relies on a multilevel buffering technique that outsmarts the simulation by making it 'see' all the hundreds of petabytes of data as being in fast memory. The buffering mechanism prevents the application from stalling when it would have needed to access data located on remote storage, allowing the application to proceed at full speed with asynchronous computing operations." KAUST's Hatem Ltaief added, “This approach also reduces the energy required to move data to and from remote storage media, which can be hundreds of times higher than the energy to perform a single computation on local memory.”
|
Lighting Up the Ion Trap
MIT News
Kylie Foy
November 4, 2020
Researchers at the Massachusetts Institute of Technology (MIT) Lincoln Laboratory have developed a compact laser light delivery system for trapped ions. The researchers used a fiber-optic block that plugs into an ion-trap chip and couples light to optical waveguides created in the chip, enabling multiple wavelengths of light to be routed and emitted to ions above the chip. MIT's Jeremy Sage said, "The integrated delivery of many wavelengths may lead to a very scalable and portable platform. We're showing for the first time that it can be done.” Added MIT's Robert Niffenegger, "Tiling these chips into an array could bring together many more ions, each able to be controlled precisely, opening the door to more powerful quantum computers."
|
Computer Model Can Predict How Covid-19 Spreads in Cities
Stanford News
Tom Abate
November 10, 2020
A team of researchers from Stanford and Northwestern universities has developed a computer model that accurately predicted Covid-19's spread in 10 major cities earlier this year by analyzing three critical infection-risk drivers. Stanford's Jure Leskovec said the model considers how persons of different demographic backgrounds, and from different neighborhoods, visit different places that are more or less crowded. Through a combination of demographic data, epidemiological calculations, and anonymous cellphone location data, the model apparently confirms that most virus transmissions are concentrated in "superspreader" sites. Leskovec said the model “offers the strongest evidence yet” that stay-at-home orders enacted this spring reduced the frequency at which people left their homes, which reduced the rate of new infections.
|
Double Patterns Could Advance Android Device Security
George Washington University
November 9, 2020
George Washington University (GWU) researchers found using multiple patterns to unlock an Android phone provides greater security than the single-pattern method, and in some cases may offer better security than four- and six-digit personal identification number unlocking used on Apple devices. The double-pattern implementation technique involves the user selecting two concurrent unlock patterns that are input in quick succession. A survey of more than 600 mobile device users found that double patterns significantly enhance the security of pattern locks against throttled attacks. Said GWU’s Adam J. Aviv, “Using two patterns to unlock an Android phone appears to provide a huge benefit for security, with little to no impact on usability.”
|
Secrets Behind 'Game of Thrones' Unveiled by Data Science, Network Theory
University of Warwick (U.K.)
November 2, 2020
Researchers from the universities of Coventry, Warwick, Cambridge, and Oxford in the U.K and Limerick in Ireland used data science and network theories to analyze George R.R. Martin's "Game of Thrones" book series. The researchers determined that the interactions between the characters in the book "A Song of Ice and Fire" were arranged similarly to those experienced in real-world relationships and interactions. The book includes more than 2,000 named characters and describes more than 41,000 interactions between them, but major characters only have an average of 150 others to keep track of, which is similar to what can be handled by the average human brain. Coventry's Joseph Yose said that "hopefully, combined with machine learning, we will be able to predict what an upcoming series may look like."
|
International Research Team Develops AI-Powered 'Electronic Nose' to Sniff Out Meat Freshness
Nanyang Technological University (Singapore)
November 10, 2020
An international team of scientists led by Nanyang Technological University, Singapore (NTU Singapore) has developed an artificial intelligence-enabled electronic nose (e-nose) that mimics the ability of mammals to evaluate meat’s freshness from its odor. The e-nose includes a "barcode" that shifts color in response to gases produced by decaying meat, and a "reader" smartphone application. The researchers trained the e-nose to identify and predict freshness from an archive of barcode colors. Its deep convolutional neural network algorithm predicted the freshness of commercially packaged chicken, fish, and beef samples with 98.5% accuracy. NTU Singapore's Chen Xiaodong said the artificial olfactory system “can be easily integrated into packaging materials and yields results in a short time without the bulky wiring used for electrical signal collection in some e-noses that were developed recently."
|
Hybrid 3D-Printing Bioinks Help Repair Knee Cartilage
New Atlas
Michael Irving
November 9, 2020
Researchers at the Wake Forest Institute for Regenerative Medicine (WFIRM) have engineered two new three-dimensional (3D) printing hybrid bioinks that can be used to print a replacement for damaged knee cartilage. The WFIRM team used the Integrated Tissue and Organ Printing System, which can print complex tissues, as well as two bioinks, to print a knee’s fibrocartilage tissue layer by layer. One bioink is a composite gellan gum and fibrinogen ink that encourages the patient’s cells to repopulate, while the other is a silk fibroin methacrylate to provide structural strength and flexibility to the printed material. Laboratory tests indicated cells could proliferate in the new material and stay viable, while the structure itself remained biomechanically sound.
|
Aveiro: Portuguese City is a Living Technology Laboratory
EuroNews
Aurora Velez
November 9, 2020
The Portuguese city of Aveiro hosts the Living Laboratory, a core hub of the Aveiro STEAM City project, engineered as a technology testbed. The project includes 16 linear kilometers (9.9 miles) of optical fibers, reconfigurable radio units, and a sensor-equipped experimental 5G network. Altice Labs’ Paulo Pereira said, “What motivates us is the realization of usage scenarios that can make sense for local government, for industry - and that represents new business opportunities.” Aveiro STEAM City also hopes to help its 34 information and communication technology firms find new talent in the science, technology, engineering, arts, and mathematics fields, and is working with the University of Aveiro to develop courses in relevant technology fields.
|
Researchers Isolate, Decode Brain Signal Patterns for Specific Behaviors
USC Viterbi School of Engineering
November 9, 2020
Researchers at the University of Southern California Viterbi School of Engineering (USC Viterbi) and New York University have developed a machine learning algorithm that isolates and decodes behaviors based on signals from the brain. USC Viterbi's Maryam Shanechi said the algorithm "can dissociate the dynamic patterns in brain signals that relate to specific behaviors one is interested in." The algorithm also can find neural patterns overlooked by other methods, as it considers both brain and behavioral signals, finding common patterns and more effectively decoding behavior represented by neural signals. USC Viterbi's Omid Sani added that the algorithm simulates common dynamic patterns between any signals, like between signals from different brain regions. Said Shanechi, "By isolating dynamic neural patterns relevant to different brain functions, this machine learning algorithm can help us investigate basic questions about [the] brain's functions and develop enhanced brain-machine interfaces to restore lost function in neurological and mental disorders."
|
|
