Association for Computing Machinery
Welcome to the June 4, 2010 edition of ACM TechNews, providing timely information for IT professionals three times a week.

HEADLINES AT A GLANCE


Hiring Expected to Be Robust in Second Half of 2010
eWeek (06/02/10) Sears, Don E.

Demand for technology professionals is rising and is expected to remain strong through the end of the year. Companies continue to hire entry-level positions, but demand for mid- and senior-level positions also is increasing. Nearly 80 percent of 800 companies polled in a new hiring survey are increasing jobs in the second half of 2010, according to Dice Holdings. "Businesses seem to be gradually loosening their grip on the hiring process as the economy improves," says Dice Holdings CEO Scot Melland. Technology job openings continue to lead the vacancies, according to a Conference Board report. "Computer and Mathematical Science occupations experienced the largest May gain, up 18,100 to 567,600, their highest level since October 2008, after a much larger April rise [of] 32,500," the report says. "The May gain reflects in part continued increases in demand for computer systems analysts and computer software engineers (applications)."


DNA Logic Gates Herald Injectable Computers
New Scientist (06/02/10) McAlpine, Katie

Hebrew University of Jerusalem (HUJ) researchers have developed DNA-based logic gates that could carry out calculations inside the body and may lead to injectable biocomputers programmed to target diseases as they arise. "The biocomputer would sense biomarkers and immediately react by releasing counter-agents for the disease," says HUJ's Itamar Willner. The logic gates are formed from short strands of DNA and their complementary strands, in conjunction with some molecular machinery, mimic their electronic equivalent. Two strands act as the input--each represents a one when present or a zero when absent. DNA computing allows calculations to be carried out in parallel, if different types of logic gates are represented by different ingredients. The HUJ team also was able to create logic gates that calculate in sequence. The HUJ system reforms after every step, enabling long sequences of calculations to be carried out. "Being enzyme-free, it has potential in future diagnostic and medical applications," says the Weizmann Institute of Science's Benny Gil.
View Full Article - May Require Free Registration | Return to Headlines


Free, Open Virtual Laboratory for Infectious Diseases
ICT Results (06/04/10)

A European research team has developed a virtual laboratory designed to help doctors match drugs to patients and make treatments more effective. The ViroLab Virtual Laboratory uses machine learning, data mining, grid computing, modeling, and simulation technologies to convert the content of millions of scientific journal articles, databases, and patients' medical histories into knowledge that can be used for treatment. "ViroLab finds new pathways for treatment by integrating different kinds of data, from genetic information and molecular interactions within the body, measured in nanoseconds, up to sociological interactions on the epidemiological level spanning years of disease progression," says University of Amsterdam professor Peter Sloot. The system continuously crawls grid-connected databases of virological, immunological, clinical, genetic, and experimental data and extracts information from scientific journal articles. The ViroLab Virtual Laboratory also could be used to create personalized drug rankings to aid in the treatment of people suffering from diseases.


Visual System Interprets Sign Languages
UAB Barcelona (06/02/10)

Researchers at the Autonomous University of Barcelona's Computer Vision Center recently demonstrated a prototype of a new visual system that interprets sign language. The visual system has been programmed to interpret more than 20 Spanish signs, which would enable deaf users to hold a basic conversation. "For them it is a non-artificial way of communicating and at the same time they can engage with people who do not speak sign language since the system translates the symbols into words in real time," says researcher Sergio Escalera. The technology consists of a video camera for recording image sequences; a computer vision and automatic learning system for detecting face, hand, and arm movement as well as any screen rolling; and a classification system for identifying each movement with a corresponding word associated with the sign. The visual interpretation system is scalable, and could be adopted for other sign languages. Precision is critical in the identification phase due to the potential variables in surroundings, such as changes in light and shadow, different physiognomies, and the speed in which signs are formed. The team focused on establishing a fixed point where individuals formed the signs and avoiding different focus points when recording.


A Community-Centric Approach to Automated Service Composition
Science in China Press (06/03/10)

The widespread diffusion of online services has enabled millions of users to voluntarily participate in the development of their own interests and benefits via service composition, but the rising numbers of services complicate the ability of users to rapidly choose and access them. The concept of "service communities" seeks to address the issue of accessing an immense and constantly shifting array of services by supplying a platform that helps users with similar goals or interests to retrieve, share, or reuse existing available solutions. Two key elements of the service community's architecture are a service pool aggregating a series of services offering identical or similar functions, and a task template capturing the access logic of some specified chore. The service community aids users through the provision of two composition styles--a process-driven style that lets users reuse existing access procedures, and a user-centric manner allowing users to iteratively access new processes in a step-by-step routine. Researchers have deployed a Web-based prototype to assess the community-centric service access concept. Participants can visually access services within Web browsers, and the user interface changes dynamically as a result of solutions from the service community.


Using Neural Networks to Classify Music
Technology Review (06/03/10) Mims, Christopher

A neural network built for image recognition is now able to classify music. University of Hong Kong students trained a conventional "kernel machine" neural network to recognize characteristics such as tempo and harmony from a database of songs from 10 genres, but discovered that the optimal number of layers of nodes needed to identify the musical genre was three. The adapted convolutional network was able to correctly and quickly identify a song with greater than 87 percent accuracy. Although the convoluted neural network was not able to identify songs outside of its training library, the team believes its ability to recognize 240 songs within two hours suggests that it is scalable. Cats, which have unique visual cortexes, served as the inspiration for the project. The Hong Kong project is the latest convoluted neural network based on a mammal to show a high level of flexibility. The results raise the question as to why such neural networks have not been used to address other problems involving perception in artificial intelligence.


Touch Is Not Enough, Say Display Experts
EE Times (06/02/10) Mokhoff, Nicolas

Experts in touch technologies recently gathered at the Society for Information Display (SID) conference to explore the future of touch and interactivity for computer and consumer appliances. Determining how people interact with machines is fundamental to finding the right mix of touch technologies. Microsoft Research's Bill Buxton says that despite the success of Apple's iPhone, much more work is still needed in touch technology. "Consumers have accepted touch technology as an input technology," Buxton says. "The question is has this acceptance stifled innovation or are others just in the mode of catching up to Apple's success." Some interface experts are encouraging people not to abandon classic tools such as pens. "The pen is natural to most people, provides accuracy for precise and detailed operation, and is efficient in content creation," says Wacom's Steve Sedaker. A paper delivered at SID called for a measuring stick for evaluating multi-touch panels. The researchers maintain that existing metrology gives useful data with regards to the mechanical and optical performances of touch panels, but tactile performance has never truly been measured.
View Full Article - May Require Free Registration | Return to Headlines


Innovative Research Reawakens Human Memories Through Intelligent Textiles
Concordia University (Canada) (06/03/10) Downey, Fiona

Prototype intelligent garments that can respond to the physical and emotional state of the wearer have been developed by research teams led by Barbara Layne of Concordia University, and Janis Jefferies at Goldsmiths, University of London. The goal of the Wearable Absence project is to develop interactive textiles that can provide wearers with personalized memories. Layne and Jefferies have combined engineered adaptors and soft cabling systems with clothing designs. The wearable system uses wireless technologies and bio-sensing devices to access a database of image and sound. The garments have wireless sensors and bio-sensing devices that enable them to record the temperature, heart rate, galvanic skin response, and rate of respiration of the wearer. The data is sent over the Internet, and a message is sent back to the clothing. Voice recordings or songs would be sent to speakers sewn into a hood or shoulder seams, and video, photographic imagery, or a scrolling text would appear on a light-emitting diode array woven into the fabric.


Students Develop Device to Help Blind Maneuver
Ben-Gurion University of the Negev (Israel) (06/03/10)

Ben-Gurion University of the Negev students recently demonstrated projects in the fields of electrical circuits and supply, microelectronics, control, communications, signal processing, computers, electro-magnetic, and electro-optics. One project involved an innovative optical radar system that acts as an assistive device for the blind. The system scans the depths of its surroundings and provides an audible alert to warn the blind of an obstacle in their path. The radar system scans its surroundings from two different angles. Elad Kuperberg and Einav Tasa, working under the supervision of professor Shlomi Arnon, incorporated a computer, two video cameras, and a scanning light source into the radar system. "This optical radar device is not only user friendly, but unlike the other solutions, it allows the blind to have the use of both of their hands," Arnon says.


The Humanities Go Google
Chronicle of Higher Education (05/28/10) Parry, Marc

Google's ambition to digitize all the world's books through its Google Books initiative is inspiring projects such as Stanford University's Literature Lab. The Literature Lab is an attempt to probe the evolution of literary style by bringing together a team of data-miners from the departments of English, history, and computer science, and have them use computer algorithms to sort, interrogate, and interpret some 1,000 digitized texts. Google has digitally archived more than 12 million books in more than 300 languages, and Stanford is hoping that projects such as the Literature Lab will give it the credibility to host a massive digital library that is being aggregated through the Google Books effort. Also vying for such a prestigious honor is the HathiTrust digital library consortium. However, Google Books has not developed the interfaces scholars require for massive digital manipulation, nor has it rigorously tagged its digitized content with metadata to ease classification and categorization. "What you may very well see is that [text-mining] becomes a more commonly accepted tool but not necessarily the center of the work of many people," says the Coalition for Networked Information's Clifford A. Lynch.


Florida State Scientists Use Unique Model to Predict Active Hurricane Season
Florida State University (06/01/10) Elish, Jill

Florida State University (FSU) researchers have developed a computer model that uses the school's high-performance computer to accurately predict hurricanes. Developed by FSU associate scholar scientist Time LaRow and colleagues at the Center for Ocean-Atmospheric Prediction Studies (COAPS), the system's first forecast last year outperformed many other models. The COAPS model uses FSU's high-performance computer to analyze huge amounts of information, including atmospheric, ocean, and land data, and is one of the few systems to use numerical models to predict hurricanes. The key component of the COAPS model is the use of predicted sea surface temperatures. The COAPS numerical model, which took five years to compile and test, requires significant computing resources to process trillions of climate calculations.


Stanford Laptop Orchestra Makes Music With Macs
San Francisco Chronicle (06/01/10) P. E1; Kosman, Joshua

Transforming computers into new kinds of musical instruments is the goal of the Stanford Laptop Orchestra (Slork) project being conducted at Stanford University's Center for Computer Research in Music and Acoustics. Professor leader Ge Wang says the underlying idea is to combine the distinctive strengths of computers and people--efficiency and networking power in the former and adaptability and creative expression in the latter. "We want to make orchestra and chamber music with these new sounds, and try to figure out how to connect people and machines musically," Wang says. Slork takes the form of a musical ensemble that uses MacBooks as instruments. Each laptop is equipped with a six-channel hub of omnidirectional speakers, creating a mix of different sounds emanating from distinct locations. Wang also created the Stanford Mobile Phone Orchestra, a mobile version of Slork that has the advantage of anytime, anywhere rehearsal and performance capability.


Follow the (Robotic) Leader
NYU-Poly (05/26/10) Hamilton, Kathleen

Polytechnic Institute of New York University professor Maurizio Porfiri has developed robots that can lead schooling fish, using advanced smart materials and mathematics. Porfiri first had to learn the mechanics of leadership in each life form before constructing a robot leader. He started with fish, and found that they make their decision on whether to school based upon what they see and the flow that they feel, which can be studied using fluid dynamics. Fish leaders beat their tails faster, mill around and accelerate to gain attention, gather a school, and lead it. Porfiri's fish robots are small enough to fit in the palm of a hand and can swim along a plane. The next developmental phase involves creating robots that can dive and surface. Porfiri hypothesized that the fish would not only accept a robotic peer that was larger than themselves but would also welcome it as a group leader, since fish of different sizes and species often school together. The robot has ionic polymers that swell and shrink in response to electrical stimulation from a battery, which propels the robot in a way that is very similar to live fish.


Abstract News © Copyright 2010 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