Association for Computing Machinery
Welcome to the October 28, 2013 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).


Researchers Draw Romantic Insights From Maps of Facebook Networks
New York Times (10/28/13) Steve Lohr

The shape of a person's social network is a powerful signal that can identify one's spouse or romantic partner, according to Cornell University's Jon Kleinberg and Facebook engineer Lars Backstrom. The researchers analyzed Facebook data from 1.3 million users, selected randomly from all users who are at least 20 years old, with between 50 and 2,000 friends, and who list a spouse or relationship partner in their profile. The data analysis resulted in about 379 million nodes and 8.6 billion links. The researchers found that the total number of mutual friends two people share is a fairly weak indicator of a romantic relationship, while a better indication is a network measure known as dispersion. High dispersion occurs when a couple's mutual friends are not well connected to one another. "A spouse or romantic partner is a bridge between a person's different social worlds," Kleinberg says. The dispersion algorithm was able to correctly identify a user's spouse 60 percent of the time. The algorithm also was able to identify people who declare themselves to be "in a relationship" 33 percent of the time.

Google Fights for Internet Freedom With New Tools
Computerworld (10/24/13) Sharon Gaudin

Google has unveiled a tool designed to provide an Internet connection to people who live in countries where governments block access. The tool, uProxy, connects people with a pathway to an online connection, via a friend in countries that do not restrict Internet access. Google unveiled the browser extension at the recent Conflict in a Connected World summit in New York City, and says it is still under development. UProxy would require a high level of trust by both parties, according to consultant Dan Olds. "The person using the connection needs to trust that his friend truly has a secure access point," he says. "And the person who is providing the connection needs to trust that the person using it isn't doing anything illegal." Google Ideas funded the research for the tool, which was developed by programmers at the University of Washington and the nonprofit Brave New Software. Google also is working to develop a real-time map of distributed denial-of-service attacks on websites around the world, in a partnership with Arbor Networks. The Digital Attack Map would let users explore historic trends and see related news reports of outages taking place on any given day.

Eliminating Unexplained Traffic Jams
MIT News (10/28/13) Larry Hardesty

Massachusetts Institute of Technology professor Berthold Horn has developed an algorithm for alleviating traffic flow instabilities that he believes could be implemented by a variation of the adaptive cruise-control systems, which uses sensors to monitor the speed and distance of the car in front of it. Horn's system also would use sensor information about the distance and velocity of the car behind it. During a simulation of traffic patterns on Massachusetts' Interstate 93, Horn found that bilateral control could be modeled using the damped-wave equation, which describes how oscillations die out over distance. He then used the Lyapunov function to demonstrate that his algorithm could stabilize the system. He says the proof accounts for several variables that govern real-life traffic flow, such as drivers' reaction times, their desired speed, and their eagerness to reach that speed. However, Horn's algorithm only works if a large percentage of cars are using it. It also requires obtaining data on a vehicle's distance and velocity, which could be possible using digital cameras. "We've developed monocular methods that allow you to very accurately get the ratio of distance to velocity," he says.

New Technique Can Find Machinery Gremlins 100 Times Faster
University of Lincoln (10/28/13) Marie Daniels

University of Lincoln researchers have developed an algorithm that can detect faults in industrial machines. The researchers say that with the improved detection and classification of faults, the time and money spent on investigating false alarms will be reduced and a machine will be operational for longer periods. The method combines two existing mathematical models for fault detection, known as a real-coded Genetic Algorithm (GA) and a K-means clustering methodology. In a GA, several possible solutions to a given problem are changed and developed toward better solutions. A GA proceeds to improve each set of solutions by repeating the application in each new generation of possible solutions until the optimum answer is found. K-means clustering is a method of analyzing a group of set objects with the goal of classifying the objects that are more similar to each other into clusters. The Lincoln researchers found that the combination of the two processes into the G3Kmeans algorithm is more effective in quickly obtaining an optimum solution, requiring only 11 repetitions to detect a specific fault. The researchers now plan to optimize their strategy by making the algorithm more wide-ranging to enable its use to detect solutions for a variety of specific applications.

100 Percent of the Image Is Restored Using a Reduced Version Containing Between 1 Percent and 10 Percent of the Information From the Original
Basque Research (10/24/13)

University of Navarre researchers have developed algorithms to reduce and optimize images. Using a reduced image with between 1 and 10 percent of the original information, the researchers say the algorithms enable 100 percent of the pixels in the initial image to be restored. "We have shown that even if we lose 100 percent of the pixels of the image, we can restore a lost image with a very high level of quality just by using the information from the reduced image," says Navarre's Daniel Paternain-Dallo. The goal of the research is to reduce the number of pixels the image contains while trying to keep all or as much as possible of the information and properties contained in the original image, Paternain-Dallo notes. He says the algorithms divide the image into small zones that are processed individually. "For each zone, we look for a value that is simultaneously the least different from all the pixels that form the zone," Paternain-Dallo says. "By following this methodology, we can design algorithms that are very efficient in terms of execution time, and capable of being adapted to the local properties of each zone of the image."

Angry Birds--The Rise of the Machines
Swiss Federal Institute of Technology in Lausanne (10/24/13)

A research team from the Swiss Federal Institute of Technology in Lausanne won an international artificial intelligence (AI) contest at the recent Beijing International Joint Conference on Artificial Intelligence with software that replicates a human-like way of playing the Angry Birds video game. The competition required that participants' machines had to solve the game's levels while only having access to the same information available to humans. The winning software was created by researchers Jason Jingshi Li, Mirko Katanic, and Arnaud Jutzeler. Their goal in the contest was the creation of a perfect intelligence that can beat anyone analyzing the game while using the same tools available to people on their smartphones. "What we're trying to do here is to understand and define intelligence in order to emulate it," Li says. The researchers say the resulting insights would be applicable to many AI disciplines, such as the processing of surveillance videos. The team used the exploitation versus exploration precept, which involves identifying which tactics to use based on the level the player is at. They first determined a way of beating every type of level the player may encounter, and then developed an algorithm tasked with choosing the strategy most likely to succeed based on previous attempts.

Why Google's Rapid Growth Means Faster Search (10/24/13) Robert Perkins

The University of Southern California (USC) has released a study showing that Google search has greatly expanded and increased in speed over the past 10 months. Google search has increased by about 600 percent the number of worldwide locations from which it serves client queries. The number of locations serving Google’s search infrastructure from October 2012 to late July 2013 climbed from less than 200 to more than 1,400, while the number of Internet service providers rose from slightly more than 100 to more than 850. Some of this growth is due to Google's reusing client networks to relay user requests and responses for search and ads. In the past, when users submitted a search request to Google, it went directly to a Google data center, whereas now the request goes to the regional network and then to the Google data center. Using the client network as an intermediary also enables lost packets to be noticed and replaced much more rapidly. The USC researchers developed a method to track and map servers that determines when they are in the same data center and where the data center is. Their next step is to quantify the performance gains for using this strategy and pinpointing underserved regions.

Sharing the Power of the Crowd
McGill University (Canada) (10/23/13) Katherine Gombay

Three years ago, McGill University researchers launched Phylo, an online puzzle game with more than 300,000 users worldwide that helped with genomic research. Now the researchers are making this crowd of players available to scientists with the goal of putting human talent to work to improve on what is already being done by computers in the field of comparative genomics. Phylo is a combination of Tetris, a Rubik’s cube, and a sliding-tile puzzle game. As gamers line up colored rectangles that represent real genetic material, they are helping to pinpoint the genetic anomalies that may be the key to a range of diseases, including diabetes, breast cancer and retinoblastoma. The researchers hope that in addition to providing solutions to genomic problems, this process also will help to promote a better general understanding of scientific research. "Our goal now is to connect thousands of scientists around the globe with hundreds of thousands of gamers," says McGill professor Jerome Waldispuhl. Phylo is currently available in 10 languages, including German, Russian, Chinese and Hebrew, and in the future, the Web site will be translated into Japanese, Arabic and Italian, according to the researchers.

Beautiful Coding--New University Project Develops Next Generation of QR Codes
University of Nottingham (United Kingdom) (10/23/13) Emma Thorne

University of Nottingham researchers have developed Aestheticodes, an initiative that aims to revolutionize embedded computer codes that enable users to access more content digitally via their smart devices. The researchers say their system designs visually appealing images and encodes them, resulting in the same interactivity of QR code, while offering a more engaging and playful experience. The technology also offers organizations the opportunity for on-brand visual interaction for products. "Graphic designers have produced a wide range of visually appealing motifs and patterns that can be used to trigger delivery of different information depending on the design and the app that reads it--for example, pointing your smartphone at the pattern on a menu in a restaurant could give you information on the special of the day," says Nottingham researcher Richard Mortier. The developers note that Aestheticodes features programming app development and graphic design services and can produce a full custom branding and interactivity service. The researchers also are working with ceramic designers to create new ceramic forms that are both aesthetically pleases and digitally trackable.

New Software Traces Origins of Genetic Disorders 20 Times More Accurately
KU Leuven (10/23/13)

KU Leuven researchers have used artificial intelligence to automate the analysis of huge amounts of genetic data, and they say their breakthrough should benefit millions of people seeking treatment for hereditary diseases. Existing analytical methods can only correctly identify genetic disorders in half of all cases, but the software suite from the team in the iMinds-STADIUS-KU Leuven group is 20 times more accurate as well as much faster. The eXtasy software suite is designed to automatically generate the most likely cause of a given genetic disorder. The researchers compare the search for disease-causing mutations in a patient's genome to looking for a specific needle in a huge haystack, considering the genomes of two healthy people can show at least 4 million differences, and most of the mutations are harmless. "EXtasy uses advanced artificial intelligence to combine whole sets of complex data into a global score that reflects how important a certain mutation is for a certain disease," says KU Leuven's Yves Moreau. "This data can consist of networks of interacting proteins, but could also include scientific publications or even scores that estimate how harmful a mutation is for the protein in question."

Killer Apps that Could Keep You Healthy
Pacific Northwest National Laboratory (10/22/13) Eric Francavilla

The Pacific Northwest National Laboratory (PNNL) recently hosted a competition in which graduate students designed two mobile apps, called FoodFeed and FL•U, to fight the threats of food-related illnesses and the flu, respectively. FoodFeed alerts users about threats that come from food, whether it is from a grocery store or restaurant. The app has three tabs, one of which is a news feed featuring articles and alerts on food recalls, illness outbreaks, and other breaking consumer-safety information. The second tab shows health code violations at restaurants. The third tab provides general information on the risks associated with food. FL•U enables users to share if they have influenza or flu-like symptoms. Users can create a customized avatar that visually displays symptoms the users submit. The app could benefit local health departments, and enable users to see how many people are sick in their area. "Whether it's a natural disaster or a disease outbreak, a public health event can come out of nowhere," says PNNL research analyst Chrissie Noonan, who helped mentor the students. "We're asking how we can stay ahead of the curve, and one answer is to develop mobile tools."

Researchers Advance Scheme to Design Seamless Integrated Circuits Etched on Graphene
UC Santa Barbara (10/22/13) Melissa Van De Werfhorst

University of California, Santa Barbara (UCSB) researchers say they have modeled an integrated circuit design scheme in which transistors and interconnects are monolithically patterned on a sheet of graphene. The researchers say the scheme could lead to ultra energy-efficient, flexible, and transparent electronics. Graphene-based transistors and interconnects are a nanoscale technology that could be used to solve issues with silicon-based transistors and metal interconnects. "In addition to its atomically thin and pristine surfaces, graphene has a tunable band gap, which can be adjusted by lithographic sketching of patterns--narrow graphene ribbons can be made semiconducting while wider ribbons are metallic," says UCSB professor Kaustav Banerjee. The researchers developed a methodology to evaluate the performance of complex circuit schemes involving many heterojunctions. "This work has demonstrated a solution for the serious contact resistance problem encountered in conventional semiconductor technology by providing an innovative idea of using an all-graphene device-interconnect scheme," says Columbia University professor Philip Kim. The proposed all-graphene circuits have achieved 1.7 times higher noise margins and 1-2 decades lower static power consumption over current complementary metal-oxide semiconductor technology. "We hope that this work will encourage and inspire other researchers to explore graphene and beyond-graphene emerging two-dimensional crystals for designing such 'band-gap engineered' circuits in the near future," Banerjee says.

Abstract News © Copyright 2013 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