Association for Computing Machinery
Welcome to the August 1, 2016 edition of ACM TechNews, providing timely information for IT professionals three times a week.

Updated versions of the ACM TechNews mobile apps are available for Android phones and tablets (click here) and for iPhones (click here) and iPads (click here).

HEADLINES AT A GLANCE


Meet the Woman Who Heads a Massively Influential All-Female Tech Team
Forbes (07/29/16) Leo King

Rochester Institute of Technology professor Vicki Hanson plans to channel her desire to enlarge the number of women entering executive positions across the tech industry into her responsibilities as the new president of ACM. "It would be nice to see more women in leadership roles at the top for-profit companies in the tech industry," Hanson says. She also sees her election to the ACM presidency as a significant event, especially as it coincided with the election of two other female computer scientists to leadership positions. "It certainly wasn't planned that the ACM would have an all-women leadership team," Hanson notes. She cites ACM's membership being only about 14-percent female as a factor, as her election signifies male members' readiness to respect women computer scientists both as equals and as the right people to lead the organization. "For young women considering career options, the election of this leadership team is a perfect opportunity to reinforce [the idea] that there really are opportunities in computing for them--all the way up to leading the industry's oldest and largest professional association," Hanson says. She still points to the career challenges women face as an underrepresented minority in the tech workforce, and she believes firms in a wide spectrum of industries can benefit by enhancing gender diversity. "We know that the best solutions to big problems are attained when diverse perspectives are brought to bear on the issues," Hanson says.


Researchers, Automakers See No Quick Path to Secure Car Networks
eWeek (07/31/16) Robert Lemos

The Automobile Industry Information Sharing and Analysis Center (Auto-ISAC) recently published its best practices for securing vehicle computer systems based on input from more than 50 automotive cybersecurity experts. The document says manufacturers should concentrate on seven security principles, including risk assessment and management, threat detection and protection, incident response, collaboration with third parties, better governance, and security awareness and training. Attempts to fortify automotive systems usually fall into one of two categories: either using cryptography to enforce behavior and trust between systems, or adding the ability to identify and mitigate an attack. Such approaches face the challenge of the protracted development cycle for vehicles, with Rubicon Labs' Rod Schultz noting, "it is very difficult when you have the massive fragmentation that you have in this industry to get any one sweeping change to happen." Meanwhile, University of Michigan researchers this month at the USENIX Security Symposium in Austin, TX, will describe a way to detect attacks using a intrusion-detection system based on knowing the timing of standard messages sent between components connected via the controller area network bus. Michigan postdoctoral researcher Kyong-Tak Cho says the method does not require manufacturers to change anything, because "it runs independently on one node that can fingerprint others and then verify and authenticate the messages."


How Tech Breakthroughs Can Shatter the Bonds of Poverty
The Huffington Post (07/28/16) Kosta Peric

Unbanked poor who cannot manage their financial resources remain trapped in a vicious circle of poverty, but digital financial services (DFS) can help break this circle by providing banking, insurance, and lending to the poor in emerging markets, writes Kosta Peric with the Bill & Melinda Gates Foundation. He says with DFS, the impoverished can manage their money digitally, storing value securely and transferring it instantly. "The global revolution in mobile communications is...enabling the poor to connect to reliable and affordable financial tools through phones, kiosks, and other digital platforms," Peric notes. However, he says the infrastructure needed to ensure universal access to DFS is still in a fledgling stage of development, and he cites three crucial technology ingredients needed to facilitate universality. Digital identity and authentication are needed to reliably guarantee new customers are who they claim to be, and one proposed solution is a "tiered know-your-customer" strategy in which more advanced services are offered as higher levels of authentication are proffered. Peric also lists distributed ledgers as an important digital financial network component, while the open sourcing of DFS technology is the third vital element. "Our foundation is looking with keen interest at ways to apply open sourcing principles to this dynamic and critically important field," Peric says.


How to Make Democracy Harder to Hack
The Christian Science Monitor (07/29/16) Scott Shackelford

With voting machines and other tools designed to uphold the integrity of democracy under threat by cybercriminals, the pressure is on to bolster the security of democratic infrastructure, writes Indiana University professor Scott Shackelford. Although regulating infrastructure is more likely to be encouraged by the appending of a "critical" label, he notes "so far, the machinery undergirding our democratic institutions has not received the same level of scrutiny as other critical infrastructure sectors." Shackelford says voting is as important an infrastructure to the U.S. as its telecom networks and financial systems. "A first step in recognizing this reality would be for the [U.S. Department of Homeland Security] to explicitly include voting booths and affiliated networks as democratic critical infrastructure, potentially as part of the already recognized 'government facilities' sector," Shackelford says. He notes such action would help clear a path for collaboration between industry and the U.S. National Institute for Standards and Technology to devise cybersecurity best practices to help jurisdictions navigate the confusing voting technology provider options. Shackelford also cites the G2 Cybersecurity Code of Conduct between the U.S. and China as an example of international cybersecurity efforts that could be expanded to include election infrastructure.


How Vector Space Mathematics Reveals the Hidden Sexism in Language
Technology Review (07/27/16)

Researchers at Boston University (BU) and Microsoft Research found rampant sexism in Google's Word2vec database, based on their queries of the vector space to find word embeddings, or relationships between words with similar meanings. The database derived from Google News texts was originally established for a neural network to identify patterns in the way words appear next to each other. BU's Tolga Bolukbasi says the sexism in Word2vec occurs because any bias in the constituent articles is captured in the geometry of the vector space. Correcting this entails searching for the opposite of gender bias while preserving the space's overall structure. The researchers quantify the nature of this warping effect by searching the vector space for word pairs that generate a similar vector to "she: he." The result is a list of gender analogies, and determining the appropriateness or inappropriateness of these analogies is a challenge met with the assistance of Amazon's Mechanical Turk. The end product is a comprehensive list of gender-biased pairs, which the researchers used to ascertain how it is mirrored in the shape of the vector space, and how this space can be changed to eliminate the warping effect. The researchers say a new list of analogies was compiled, rated by Mechanical Turk, and converted into a less-biased vector space.


Body Talk: A New Crowdshaping Technology Uses Words to Create Accurate 3D Body Models
Max Planck Society (07/26/16)

Researchers at the Max Planck Institute for Intelligent Systems and the University of Texas at Dallas have developed a crowdshaping technology that creates accurate three-dimensional (3D) body models from two-dimensional photos using crowdsourced linguistic descriptions of body shape. The Body Talk system takes a single photo and produces 3D body shapes that resemble the person and are accurate enough to size clothing. The system works with 15 volunteers who rate the body shape in the photo using 30 words or less, and employs machine learning to determine the relationship between verbal descriptions of bodies and their actual 3D shapes. The researchers found that realistic 3D bodies could be created using as few as 10 words. They also developed a "3D paparazzi" application, which takes a single photo of a celebrity, has it rated by 15 people, and then creates a 3D avatar of the celebrity that can be animated. In addition, the system can extract measurements from the crowdshaped bodies with an accuracy sufficient for many clothing-sizing applications. "This system can be used in many areas of science and medicine, for example in studies of body perception disorders, obesity, or cross-cultural attitudes about body shape," says Max Planck Institute researcher Michael J. Black. The researchers presented the Body Talk system at the ACM SIGGRAPH 2016 conference in Anaheim, CA.


Top Programming Languages Trends: The Rise of Big Data
IEEE Spectrum (07/26/16) Nicholas Diakopoulos

Big data programming languages are gaining in importance because they enable the mining of massive datasets, the collection of which across virtually all sectors of government, science, and commerce has become the norm. Google's open source Go language, originally created to address the company's issues with scaling systems and concurrent programming, has had the most traction. Currently in 10th place in IEEE Spectrum's rankings, Go has risen 10 positions in only two years, mainly thanks to the large increase in related activity on the GitHub source code archive. Also having risen significantly since 2014 is current fifth-place-holder R, which registered about 46 percent more questions on Stack Overflow, and increasingly is cited in scholarly research papers. Meanwhile, there has been a decline in use of proprietary data-analysis languages such as Matlab and SAS, although usage of both languages is continuing to expand at slower rates than some of the languages that are displacing them. Java and Python still dominate in terms of jobs, but recruiter interest in R and Scala has climbed substantially since 2014. Still, IEEE Spectrum found about 15 times as many job listings for Python developers as for R programmers. R may offer exceptional visualization and exploratory analysis and is popular with academics writing research papers, but Python has significant benefits for users in production environments.


White House Celebrates Cyber Contests to Attract Young Talent
FedScoop (07/28/16) Jeremy Snow

The White House on Wednesday hosted the Cybersecurity Competitions Workshop to celebrate cyber education and the benefits of competitive cyber competitions for a new generation of cybersecurity experts. The event was organized by the U.S. Office of Science and Technology Policy (OSTP) and included students, organizers, and sponsors from previous competitions. CyberPatriot VIII was held in April and featured more than 3,300 student teams competing to patch mock vulnerabilities. The Collegiate Cyber Defense Competition and the U.S. Cyber Challenge also tested the cybersecurity skills of student participants. OSTP aims to encourage more students to enter the cyber workforce to meet a growing demand for cybersecurity experts. These contests also are part of an effort to promote higher diversity in the cyber talent pool, and this year's CyberPatriot saw the proportion of female participation rise from 6 percent to 23 percent. "Given the nation's cybersecurity workforce needs, we cannot afford to miss out on any dimension of the talent pool," notes a White House blog entry. "Competitions give students from all geographic locations and walks of life opportunities for awareness about cybersecurity as a vocation and an introduction to the subject matter."


Machine Learning Researchers Team Up With Chinese Botanists on Flower-Recognition Project
Phys.org (07/27/16) Guobin Wu

Microsoft Research Asia scientists, working with botanists at the Institute of Botany, Chinese Academy of Sciences (IBCAS), have developed the Smart Flower Recognition System, which is designed to quickly identify any flower or plant. The researchers trained a deep neural network to recognize images using a set of learnable filters. Each filter is initially convolved across the width and height of the input volume, computing the dot product between the entries of the filter and the input. The process produces a two-dimensional activation map of that filter, and the system then learns filters that activate specific types of features at a given spatial position in the input. After entering millions of pictures into the deep-learning framework, the researchers say they enabled the engine to accurately identify images more than 90 percent of the time. "The flower-recognition engine enables domain experts to acquire plant distribution in China in an efficient way," says IBCAS researcher Zheping Xu. The researchers say the next step is to create applications based on the flower-recognition engine so botanists can conduct their research more efficiently.


Researchers Are Designing Data-Exchange Platform That Can Act as City's Dashboard
A*STAR Research (07/27/16)

The A*STAR Urban Systems Initiative (USI), part of Singapore's Smart Nation initiative, is refining the art of big data consolidation in the cityscape through programs unified by the A*STAR Data Analytics Exchange Platform (A*DAX). USI director Ng See-Kiong says A*DAX is an open-standards, cross-domain platform for data connectivity, sharing, and analytics throughout Singapore's public and private sectors. "The platform itself is a controlled environment with strict user access controls--if you are the data owner or service provider, you can give access to other users," Ng says. Once a business agreement had been reached between the data buyers and sellers, the platform designers had to guarantee the different types of data could be fused and extracted concurrently. Ng's team created A*DAX Fusion middleware, enabling users to issue a single query to any combination of multiple datasets of different data types based on a framework called representational state transfer application program interfaces. A*DAX's analytic capabilities needed refinement, because "users didn't want more data--they wanted more answers," Ng says. Instead of giving away the raw data, Ng says the team designed the platform to "provide people with a way to query it for less specific information, but more valuable insights such as aggregates or predictions." To better predict human behavior in urban systems, the USI's Complex Systems Arm supplies a mechanism-based modeling and simulation architecture to complement the data, visualization, and analytics from other USI programs.


Ludwig the Robot Designed to Help Alzheimer's Patients
CBC News (Canada) (07/26/16) Errol Nazareth

A walking, talking robot called Ludwig developed by researchers at the University of Toronto (U of T) is envisioned as a tool to help people suffering from Alzheimer's disease and other forms of cognitive decline. Ludwig has attracted much attention and praise since its unveiling at a retirement home in Toronto. U of T professor Frank Rudzicz notes robotics is finding increasing use in healthcare environments. He says the straightforward, simple, and entertaining interaction of robots with autistic children has yielded promising results, which his team hopes to repeat with adult Alzheimer's patients. The team initially is concentrating on speech recognition and Ludwig understanding and responding appropriately. However, the researchers also are investigating how the robot can sense changes to a person's cognitive state based on their speech patterns. Further out is a project to make Ludwig capable of detecting "when someone's confused or not interested in conversation," Rudzicz says. "Over next three months, we'll be getting him to talk to residents and trying to figure out what ways he talks are most useful to a free-flowing conversation," he says. "When we learn that, we'll build that into the robot."


Ultra-Flat Circuits Will Have Unique Properties
Rice University (07/25/16) Mike Williams

Researchers at Rice University have analyzed hybrids that put two-dimensional (2D) materials such as graphene and boron nitride side by side to see what happens at the border. They found the electronic characteristics of such co-planar hybrids differ from bulkier components. The researchers say 2D interfaces created "a highly nonlocalized charge transfer"--and an electric field along with it--that greatly increased the junction size. They say this could give them an advantage in photovoltaic applications such as solar cells. The lab of Rice's Boris Yakobson built a computer simulation of a hybrid of graphene and molybdenum disulfide and also considered graphene-boron nitride and graphene in which half was doped to create a positive and negative (p/n) junction. The team's calculations predicted the presence of an electric field should make 2-Schottky, or one-way, devices such as transistors and diodes more tunable based on the size of the device itself. Yakobson says there is no reason why they cannot build 2D rectifiers, transistors, or memory elements, "but unless we develop a proper fundamental knowledge of the physics, they may fail to do what we design or plan."


Abstract News © Copyright 2016 INFORMATION, INC.
Powered by Information, Inc.


To submit feedback about ACM TechNews, contact: [email protected]
Current ACM Members: Unsubscribe/Change your email subscription by logging in at myACM.
Non-Members: Unsubscribe