Welcome to the June 2, 2010 edition of ACM TechNews, providing timely information for IT professionals three times a week.
HEADLINES AT A GLANCE
Chinese Supercomputer Is Ranked World's Second-Fastest, Challenging U.S. Dominance
New York Times (05/31/10) Markoff, John
China's Dawning Nebulae supercomputer, based at the National Supercomputing Center in Shenzhen, has been ranked as the world's second-fastest machine, passing European and Japanese systems. Dawning Nebulae achieved a sustained computing speed of 1.27 petaflops, which makes it the world's fastest in terms of theoretical peak performance, but that is considered a less significant measure than the actual computing speed achieved on a standardized computing test. The Cray Jaguar, based at the Oak Ridge National Laboratory in Tennessee, remains the world's fastest supercomputer, according to the latest semiannual supercomputer ranking. Dawning Nebulae is based on chips from Intel and NVIDIA. China also is developing a new system based on a microprocessor designed and manufactured in China, which is expected to be complete later this year. "I wouldn’t be surprised if by the end of this year they surpass the scientific computing power of the [European Union] countries combined and have a computer system with an achieved performance to reach the No. 1 position on the Top500," says University of Tennessee computer scientist Jack Dongarra.
Impossible Figures Brought to Life in Virtual Worlds
New Scientist (06/02/10) Marks, Paul
Chinese University of Hong Kong (CUHK) computer scientists have developed software that depicts physically impossible images in three-dimensional virtual environments. The software deforms the impossible figures by a different amount for each viewing angle to maintain the illusion of impossibility. "We model the shape as if it's a non-rigid body, like an amoeba, which can be deformed," says CUHK's Tai-Pang Wu. The system could be applied to graphics packages for multimedia artists, Wu says. "People love visual illusions and tricks and this is certainly an interesting geometric exercise," says Tel Aviv University's Daniel Cohen-Or.
The World: A Global Village Called Babel
ICT Results (06/01/10)
European researchers working on the Statistical Multilingual Analysis for Retrieval and Translation (SMART) project have developed technology that will enable machine translation using statistical analysis. SMART researchers were inspired by the Pascal Network of Excellence, which sought to develop cooperative ties among Europe's leaders in pattern analysis, statistical modeling, and computational learning. "We sought to develop more effective statistical learning methods, apply them to machine translation, and then prove the platform through rigorously measured case studies," says SMART coordinator Nicola Cancedda. SMART case studies focused on computer-aided translation (CAT) and cross-language information retrieval (CLIR). CAT is used by professional translators, while CLIR is used to acquire information from a foreign language document. "In our case study, the SMART platform increased words per hour by 5 to 40 percent," Cancedda says. The SMART project also developed real-time learning tools that can "teach" software new terms and translations. "We have developed tools where the software learns all the time, so it becomes much, much better over time, and so much more valuable," he says.
Surveillance Software Knows What a Camera Sees
Technology Review (06/01/10) Simonite, Tom
University of California, Los Angeles (UCLA) researchers have developed Image to Text (I2T), a computer vision system that can generate a real-time text description of what is happening in a surveillance camera video feed. The researchers put a series of computer vision algorithms into a system that accepts images or video frames as input and generates summaries of the input. I2T uses an image parser to break down an image by separating the background from objects in the picture. Next, the meaning of the objects is determined. "This knowledge representation step is the most important part of the system," says UCLA professor Song-Chun Zhu. I2T includes a database of more than two million images containing objects that have been identified and classified into more than 500 categories. The video-processing system uses algorithms that can describe the movement of objects in successive frames. Although the system demonstrates a step toward what Zhu calls a "grand vision in computer science," I2T is not ready for commercialization. Improving the system's knowledge of how to identify objects and scenes by adding to the number of images in the database will help I2T grow, Zhu says.
Software Developers Tackle Child Grooming on the Net
Lancaster University (06/01/10)
Lancaster University (LU) computer scientists are developing a tool that could help law enforcement agencies catch adults in chatrooms posing as children by determining a person's age and gender using language analysis techniques. The researchers say the software also could be used to identify the stylistic footprints of pedophiles. Once fully developed, the language analysis techniques will be automated, potentially freeing up police time and adding to the expertise already deployed within investigations that identify and locate child sex offenders. "Pedophiles often pose as children online and our research indicates that children don't find it easy to spot an adult pretending to be a child," says LU professor Awais Rashid. "We hope to develop an automated system which can pick up on quirks of language particular to a certain age group."
House Allocates $84 Billion for Technology R&D
InformationWeek (06/01/10) Montalbano, Elizabeth
The U.S. House of Representatives has passed the America Competes Reauthorization Act of 2010, which provides $85 billion in science and technology research and education funding. The act, first enacted in 2007, is backed by the Obama administration and is designed to help keep the United States competitive in science and technology. It provides funding over five years to the National Science Foundation, the Department of Energy's Office of Science, the National Institute of Standards and Technology, and other organizations. In addition to promoting research and development funding, the administration also is fostering science, technology, engineering, and mathematics education for children with its "Educate to Innovate" campaign, a partnership of the federal government and the private sector.
New Automated Tool 'Debugs' Nuclear Weapon Simulations
Purdue University News (06/01/10) Venere, Emil
Researchers at Purdue University and Lawrence Livermore National Laboratory have developed AutomaDeD, an automated program designed to debug the simulations used to certify nuclear weapons. "The simulations take several weeks to run, and then they have to be debugged to correct errors in the code," says Purdue professor Saurabh Bagchi. The simulations, which can have as many as 100,000 lines of code, must accurately show reactions taking place on the scale of milliseconds. "If things start going wrong, AutomaDeD would stop and flag which process and which part of the code in the process is likely anomalous," Bagchi says. AutomaDeD works by splitting a simulation into several windows of time, called phases. The researchers say that, during testing, AutomaDeD was 90 percent accurate in identifying the time phase when stalls and hangs occurred, 80 percent accurate in identifying the specific tasks that were the sources for stalls and hangs, and 70 percent accurate in identifying the interference faults. The same technique could be used to find errors in other parallel applications, such as those used for climate modeling and high-energy particle physics.
Classic Grammar Model Can Be Used for Computerised Parsing
University of Gothenburg (Sweden) (05/31/10)
University of Gothenburg researcher Kenneth Wilhelmsson has developed parsing software for interpreting the grammatical structure of modern Swedish text. "The grammatical analysis in the program is performed mostly at the main clause level, which can be seen as a big advantage, as the task is then less complex but still gives usable results," Wilhelmsson says. The program is designed to conduct a series of steps in identifying complex sentence elements, rather than conducting the entire analysis in one attempt. Language researchers can use the program to search for different grammatical phenomena. The program also has been used to build a system that automatically generates queries from a Swedish text. The program accesses the Swedish Wikipedia database, and can generate queries when a text is opened. "This is intended as an alternative to most other modern query programs where the user cannot know whether a query can actually be answered by the knowledge base at all, and where variations in the formulation of the query may mean that information that is there is missed," Wilhelmsson says.
Defence Lab Reveals Ultimate CCTV
BBC News (05/28/10) Gill, Victoria
The United Kingdom's Defence Science and Technology Laboratory (DSTL) is developing a package of surveillance systems that can recognize insurgents or terrorists. The DSTL researchers are developing high-resolution imaging technology with built-in software to detect and follow insurgents based on their behavior. "You can't tell who is an insurgent by what they look like, but you can track their behavior," says DSTL chief technologist Andrew Seedhouse. The surveillance techniques will be combined with software that can detect unusual patterns in behavior, such as two vehicles meeting in a concealed location. "It's like having the ultimate [closed-circuit TV] system with super sensors on the ground and flying round in the air," Seedhouse says. The package also will include a 128-megapixel high-resolution camera that can capture a four-kilometer by four-kilometer area. Another system, called a hyperspectral camera, can analyze color to distinguish a camouflaged vehicle from the vegetation it is hidden within. Seedhouse says that all of the systems use the insurgents' activities against them. "We will get to the point where we can prevent them from hiding in the background noise and where they become a target," he says.
Project Aims to Halve Cost of a Data Center
InfoWorld (06/02/10) Nystedt, Dan
Taiwan's Industrial Technology Research Institute (ITRI) plans to create Container Computer 1.0, a set of standards for building data centers inside 20-foot shipping containers. ITRI researchers say the standardized container data center will cost about half of today's systems, will require less energy, and will be easier to use. "We hope to have a prototype available by the end of the year," says ITRI researcher Chiueh Tzi-cker. The ITRI initiative aims to make containerized data centers a standardized product, similar to the way standardization has cut the price of personal computers. The researchers also plan to create Cloud Operating System 1.0, an operating system for the data centers that also will include a package of third-party programs.
New System Using Bacterial Communities to Solve Complex Problems
Universidad Politecnica de Madrid (Spain) (05/31/10) Martinez, Eduardo
Researchers at Universidad Politecnica de Madrid have developed a system that uses bacterial communities to autonomously solve complex problems. The system represents another step in the development of synthetic biology and bacterial computing. The system's algorithms help to synchronize different bacteria according to the bacteria's natural capabilities. The system has been used in computational applications such as autonomous complex problem solving by bacterial communities, and in the design of a population oscillator modeled after the likeness of a client/server architecture. The researchers say the system also could be applied to medicine and ecology, since it has been validated both at the biological and computational levels.
A Novel 'Cuckoo Search Algorithm' Beat Particle Swarm Optimization in Engineering Design
Xin-She Yang, a research scientist at the National Physical Lab in the United Kingdom, and Suash Deb of the C. V. Raman College of Engineering in Bhubaneswar, India, have significantly improved their search algorithm technique for designing engineering structures. The algorithm is based on the behavior of the female cuckoo bird, which lays fertilized eggs in the nest of another species with hopes that her brood would be raised by the surrogate parents. Yang and Deb created a mathematical formula for the principles of the cuckoo bird's reproduction strategy, and then converted them to an algorithm. The software is designed to treat the various design parameters and constraints as "eggs," discard some due to a lack of fitness, and then send the successful solutions along for additional rounds of tests until it finds the optimal solution. Although tests showed the cuckoo search technique was more efficient than particle swarm optimization, the researchers believe they could be combined to produce an even more powerful approach to engineering design.
Detecting a Crime Before It Happens
Los Angeles Times (05/28/10) Drogin, Bob
Government scientists at the Homeland Security Advanced Research Project Agency are immersed in research to see whether various high-tech devices can be exploited to spot nonverbal cues from people with malicious intent in the hope of preventing crimes. The researchers have connected high-resolution cameras, low-level lasers, and other devices to measure fidgeting, heart rate, pupil dilation, skin temperature, and other cues. The sensors would record key data as each traveler moved down a security line, and a computer algorithm would then analyze any changes triggered by a guard's questions and raise an alert if necessary. Meanwhile, Homeland Security's human factors branch has spent close to $20 million on experimentation with micro-expressions, or rapid facial muscle twitches, that may or may not signal hostile intent. There is a clear need for improvement, as demonstrated by Homeland Security screeners' failure to detect travelers who were later connected to failed terrorist plots in New York and Virginia, jihadist training in Pakistan, and deadly incidents in Afghanistan, India, and Somalia, according to a Government Accountability Office report.
Video Game Accessibility Project to Help Blind Children Exercise
University of Nevada, Reno (05/28/10) Wolterbeek, Mike
University of Nevada, Reno (UNR) researchers have developed motion sensing-based tennis and bowling video games that use physical activity as input and might be useful in the fight against obesity. VI Fit is designed to help visually impaired children become more physically active and healthy through video games. "Lack of vision forms a significant barrier to participation in physical activity and consequently children with visual impairments have much higher obesity rates and obesity-related illnesses such as diabetes," says UNR professor Eelke Folmer. "Our games are adaptations of the popular Nintendo Wii Sports exercise games that have been modified so they can be played without visual feedback." VI Tennis uses the gameplay of Wii Sports tennis to provide audio and vibrotactile cues that indicate when to serve and when to return the ball. "We found our game to engage children into levels of active energy expenditure that were high enough to be considered healthy, which shows the feasibility of using video games as a health-intervention method," Folmer says. VI Bowling uses a motor-learning feature that allows players to find the direction in which to throw the ball using vibrotactile feedback.
Abstract News © Copyright 2010 INFORMATION, INC.
To submit feedback about ACM TechNews, contact: [email protected]
Current ACM Members: Unsubscribe/Change your email subscription by logging in at myACM.