Welcome to the February 4, 2011 edition of ACM TechNews, providing timely information for IT professionals three times a week.
Also, please download our new ACM TechNews iPhone App from the iTunes Store by clicking here and our new ACM TechNews iPad App by clicking here.
HEADLINES AT A GLANCE
Future Surgeons May Use Robotic Nurse, 'Gesture Recognition'
Purdue University News (02/03/11) Emil Venere
Purdue University researchers are developing a robotic system that can recognize hand gestures during a surgical procedure. The system would be used to control a robotic scrub nurse or to tell a computer to show medical images of the patient, both of which could help reduce the length of procedures and the risk of infection, says Purdue professor Juan Pablo Wachs. He says the vision-based hand gesture recognition technology could have other applications, such as the coordination of emergency response activities during disasters. The system uses algorithms and a camera to identify hand gestures as commands to instruct the robot or computer. "You want to use intuitive and natural gestures for the surgeon, to express medical image navigation activities, but you also need to consider cultural and physical differences between surgeons," Wachs says. He says one challenge for the technology is being able to understand context and differentiate intended gestures from unintended ones. The gesture algorithms are based on anthropometry, which involves predicting the position of the hands based on the position of the head. The researchers also are developing ways "to anticipate what images the surgeon will need to see next and what instruments will be needed," Wachs says.
Can Focus on Video Games and Visual Effects Enhance STEM Education Efficiency?
International Business Times (02/02/11)
Video games and visual effects could be used to improve science, technology, engineering, and mathematics (STEM) education in the United Kingdom, according to a new report from the National Endowment for Science, Arts, and Technology. The suggestion is one of 20 recommendations made by the report, which was based on a review by Square Enix's Ian Livingstone and Double Negative's Alex Hope. The report says that schools continue to focus more on boosting the office skills of students rather than providing them with rigorous computer science and programming skills. It also says that informing students of the opportunities in high-tech industries such as video games and visual effects could help make STEM more exciting, attract more students to the subjects, and help students understand that computer science and programming skills can lead to creative careers. "We need to set in motion a virtuous circle where video games and visual effects help draw young people into maths, physics, and computer science, and improve their learning outcomes, in turn enlarging the talent pool for these industries in the future," the report says. "Schools should do more to encourage cross-curricular learning."
Crowd Workers Are Not Online Shakespeares, but HCII Researchers Show They Can Write
Carnegie Mellon News (PA) (02/02/11) Byron Spice
Researchers at Carnegie Mellon University's Human-Computer Interaction Institute have developed CrowdForge, a crowdsourcing framework that breaks down complicated problems into simple, independent tasks that can be finished quickly and inexpensively. To develop CrowdForge, the researchers conducted experiments in which volunteers completed one task in the process of writing an article, such as creating an outline, conducting research, or writing the sentences. CrowdForge was used to write five articles that included 36 subtasks each. Individuals were hired to write similar articles. The researchers, led by professor Aniket Kittur, found that the crowdsourced articles were rated higher than those produced by individuals. "This is exciting because collaborative crowdsourcing could change the future of work," Kittur says. "We foresee a day when it will be possible to tap into hundreds of thousands or millions of workers around the globe to accomplish creative work on an unprecedented scale." CrowdForge utilizes Amazon's Mechanical Turk, which is used by employers for outsourcing tasks such as identifying objects in photos, writing product descriptions, and transcribing audio recordings.
Give Your Dashboard the Finger
Technology Review (02/02/11) Duncan Graham-Rowe
Drivers will be able to control the dashboard without taking their hands off the steering wheel or their eyes off the road using a new gestural interface developed by researchers at the German Research Center for Artificial Intelligence (DFKI). Geremin, an experimental in-car interface, uses several sensors to detect the movement of a driver's right index finger as it disrupts an electric field. The system has the potential to interpret a wide range of commands, and would be able to offer drivers more functions and applications than cars currently offer via buttons and controls, says DFKI's Christian Muller. In tests, Geremin was able to distinguish 10 gestures, such as moving the finger up or down, left or right, or tracing out circles, triangles, and squares, with an accuracy rate of 86 percent. Geremin would be less expensive than installing cameras in cars to monitor driver movements, Muller says. The team has plans to significantly extend the gesture set for Geremin. "We will combine this with speech recognition in order to allow people to dictate text messages in the car," Muller notes.
Stephen Fry Backs Project to Make Web More Accessible
University of Southampton (United Kingdom) (02/01/11) Joyce Lewis
The recently launched Fix the Web project seeks to make the Web more accessible for disabled and older users. Fix the Web provides options for reporting accessibility issues to Web sites through a form on the Fix the Web site, using Twitter, email, or the ATBar toolbar. Developed by researchers at the University of Southampton, ATBar offers a Fix the Web button that quickly launches a report form, which volunteers take through a checking process and send to Web site owners with information about the accessibility issue. The ATBar incorporates text resize, text to speech, style, and reference-setting buttons. Thus far, Fix the Web has reported 388 sites, 20 sites have been fixed and many more are in progress, and has 296 volunteers. The researchers say that ATBar has received more than 3 million toolbar hits over the past six months.
Turning Reviews Into Ratings
MIT News (02/03/11) Larry Hardesty
Researchers at the Massachusetts Institute of Technology's Computer Science and Artificial Intelligence Laboratory's Spoken Language Systems Group have developed a system that automatically analyzes user reviews for information that is not commonly found on other Web sites. The system's algorithm was originally developed as a part of speech-recognition technology, and has now been adapted to determine the grammatical structure of the sentences that make up reviews. Many of the review sites also provide a numerical rating system to accompany the verbal reviews. The system, developed by research scientist Stephanie Seneff and graduate student Jingjing Liu, uses a second group of algorithms to apply the numerical ratings to the meaning of the written reviews. For example, a review of "excellent" commonly receives a five when using a five-star rating system, while a review of "horrible" often receives just one star. The system can then determine new words that have a positive or negative connotation, based on the results of the first word-to-rating analysis. Microsoft's researcher Xiao Li says the researchers' work is distinct because "they do a lot of linguistic analysis."
Checkout AI Uses Camera to Tell Your Apples Apart
New Scientist (02/04/11) Yuriko Nagano
Toshiba researchers have developed a system composed of a Webcam, image-recognition software, and machine-learning programs, with the goal of improving the grocery store self-checkout process by identifying loose goods. The system can differentiate between products that look almost identical by recognizing slight variations in color, shape, and small markings on the surface. The system compares an image of the product taken with a Webcam to a database of images. The software provides a list of products that are similar, with the closest match listed first. If the product chosen from the database matches the item, the user presses a button to complete the purchase. If the system fails to find the correct match for an item, users can help train it by choosing the correct product name and picture to match it. "This system gets smarter as you use it more," says Toshiba researcher Susumu Kubota. In tests, the system was able to recognize produce even if it was placed in a clear plastic bag, Kubota notes. However, the system still has some hurdles to overcome before it is ready for commercial use, such as dealing with dishonest customers and perfecting a system that recognizes generic objects, which is much more difficult than developing facial-recognition technology, says Keiji Yanai of the University of Electro-Communications.
EU Warns It Lags Behind in Global Innovation Race
Associated Press (02/01/11) Raf Casert
The European Union's (EU's) scientific and industrial innovation efforts lag far behind the United States and Japan, prompting the EU Commission to call for measures to promote private investment and greater business engagement in the application of research results. A series of research and innovation indicators for the EU show that the United States was overtaking Europe by 49 percent in 2010, up from 45 percent three years earlier. Meanwhile, Japan's lead climbed from 32 percent to 40 percent between 2006 and 2010. Among the indicators analyzed by the Innovation Union scoreboard are license and patent revenue, new doctoral degrees, and business research and development (R&D) expenditures. "The EU share of world R&D expenditure has decreased by a fifth over 15 years," says EU science commissioner Maire Geoghegan-Quinn. The EU's slice of the world R&D expenditure declined from 28.6 percent in 1995 to 23.4 percent in 2008, while at the same time Asia grew its share from 21.9 percent to 29.7 percent. "If we are going to grow our way out of the present economic crisis, innovation, investment in research, education, and technology is the way that we have to do that," Geoghegan-Quinn says.
Mobile Phone Software to Help Keep Kids Safe
Lancaster University (02/01/11)
Researchers at Lancaster University spinoff company Isis Forensics have developed Child Defense, mobile phone software that enables children to find out more information about the people they are communicating with online. Child Defense uses language analysis technology to identify terminology that is specific to certain age groups. The software can link with social networking Web sites such as Facebook and Twitter, enabling users to develop profiles of the people they are talking to on their phones. In testing, Child Defense was as effective as server-based programs in determining if an adult is masquerading as a child. "Using state-of-the-art language analysis software it gives children a powerful tool which can help them work out who they are really talking to online," says Isis Forensics' James Walkerdine. "Parents in our focus groups told us they would much prefer to see software solutions that empowered and educated their children to help them protect themselves." The software will be made freely available for use on iPhone, Google, and Nokia phone apps.
New IU Malware Records Data From Cell Phones
Indiana Daily Student (02/01/11)
Indiana University researchers have developed Soundminer, a smartphone Trojan Horse virus that can record and steal conversations, keypad sounds, credit card numbers, and bank account information. The researchers developed the virus with the goal of improving security for Android-based smartphones. "We're in the business of building secure systems," says Indiana professor Apu Kapadia. "We want to live in a more secure world, but part of that job is also trying to be one step ahead of the bad people." The researchers developed Soundminer by tracking trends in smartphone usage and researching commonly known threats. "If you can show that there's some security flaws in Android and those can be fixed, that will actually benefit a large number of people," says City University in Hong Kong visiting researcher Roman Schlegel. Since smartphones are just like small computers, protecting them from malware is mainly about remaining aware of new viruses and new defense strategies, says Indiana's Scott Wilson.
Taking Unpleasant Surprises Out of Cosmetic Surgery
American Friends of Tel Aviv University (02/01/11)
Tel Aviv University researchers have developed software that can provide cosmetic surgery patients with more accurate before-and-after pictures to help them get a realistic idea of what they will look like after the surgery. The software deals with the computer modeling problem known as predicting deformations of non-rigid objects. The tool retrieves geometric objects similar to the way Google finds Web pages. In addition to helping patients, the program also can help surgeons achieve the best possible outcome for the patient. "Our program is more like a virtual mirror," says Tel Aviv's Alex Bronstein. "It gives surgeons and their patients a way to see a [three-dimensional] before-and-after image as though the patient has really undergone the operation." The researchers developed the program by analyzing data from former plastic surgery patients and took several variables into account, including the patients' ages and different tissue types. The program was fed numerous pre- and post-surgery images so the computer could learn how to produce more accurate post-surgery images.
A Matter of Timing: New Strategies for Debugging Electronics
University of Wisconsin-Madison (02/01/11) Sandra Knisely
University of Wisconsin-Madison professor Azadeh Davoodi is studying solutions for timing errors, which are electronic bugs that can occur after a chip is fabricated and can cause components to slow and take longer to execute operations. Davoodi says that due to their small size and large number of components, chips cannot realistically be completely void of bugs before fabrication. "The nanoscale components in the chip are so small they can have weird physical behaviors that can only be detected after they are fabricated," he says. Timing errors often are only revealed when specific operations are performed together, which makes testing tedious and difficult, because testers have to predict how a chip will be used. Davoodi's team is developing sensor components that can be built into the chip and will provide timing information, allowing manufacturers to solve errors faster. "We want to increase the timing observability inside the chip," he says.
Simulating Worst-Case Scenarios
Johns Hopkins Gazette (01/31/11) Mark Guidera
Johns Hopkins University recently launched the Center for Advanced Modeling (CAM) in the Social, Behavioral and Health Sciences, which will focus on advanced computer simulation and modeling technology. "With the launch of this center, Johns Hopkins is firmly planting its flag in the ground and saying we are going to be a mecca for groundbreaking research and applied work in the field of agent-based modeling," says Johns Hopkins professor Joshua M. Epstein. He says the center will unite some of the U.S.'s leading experts in fields such as emergency medicine, disaster health, social behavior, supercomputing, and economics to advance the agent-based modeling field. The agents used in CAM simulations are programmed to respond to real or imagined threats in the same way that actual people would. Epstein says CAM simulations could help predict how societies will react to disasters such as infectious disease outbreaks, toxic chemical spills, or natural disasters. "I see this as a place where the top professors and researchers from around the country, indeed the world, will want to come and work on collaborative projects, participate in symposia or develop entirely novel lines of research," he says.
Abstract News © Copyright 2011 INFORMATION, INC.
To submit feedback about ACM TechNews, contact: [email protected]
Current ACM Members: Unsubscribe/Change your email subscription by logging in at myACM.