Carnegie Mellon Algorithm Identifies Top 100 Blogs For News

Being among the first to pick up on Internet news and gossip and rapidly detecting contamination anywhere in a water supply system are similar problems, at least from a computer scientist’s point of view. Both can be solved with a versatile algorithm developed by Carnegie Mellon University researchers.

Using a problem-solving method called the Cascades algorithm, Carlos Guestrin, assistant professor of computer science and machine learning, and his students compiled a list of the best 100 blogs to read to find the biggest news on the Web as early as possible. It includes well-known blogs, such as Instapundit and Boing Boing, but also some more obscure ones like Watcher of Weasels and Don Surber.

“The goal of our system when looking at blogs is to detect the big stories as early on and as close to the source as possible,” Guestrin said. He, Andreas Krause and Jure Leskovec, doctoral students in computer science and machine learning, respectively, analyzed 45,000 blogs (those that actively link to other blogs) to compile the list, checking the time stamps to determine where news items were being posted first.

But reading even 100 blogs, many of them with numerous postings, may be more than many Web surfers can handle. Recasting the problem, the researchers used their algorithm to compile a list of blogs if a person wanted to read only 5,000 postings. This list is quite different, with “summarizer” blogs, such as The Modulator and Anglican predominating.

Similarly, Guestrin and his students used the same algorithm to determine the optimal number and placement of sensors for detecting the introduction and spread of contaminants in a municipal water supply. Their report on the blog and water system case studies, “Cost-Effective Outbreak Detection in Networks,” was presented at the Association for Computing Machinery’s International Conference on Knowledge Discovery and Data Mining earlier this year.

“Nothing demonstrates the versatility of Carlos’ algorithm better than its ability to solve these two difficult and seemingly different problems,” says Randal Bryant, dean of Carnegie Mellon’s School of Computer Science. “It’s a credit to Carlos’ insight and inventiveness, but also a testament to the power of computational thinking. Computer scientists increasingly are developing common methods for solving problems that apply across any number of disciplines.”

Guestrin began work on the Cascades algorithm in 2004 to find a way to balance the cost of collecting information with the need for collecting the information early and close to its source. Initially, this addressed problems in designing wireless sensor networks — a technology that potentially can monitor such important conditions as water quality, building temperature, vital signs of nursing home residents, algal blooms in lakes and the structural integrity of bridges. In all of these cases, deploying the wrong number of sensors or putting them in the wrong places wastes money and produces poor information.

The algorithm allows for near-optimal placement of sensors by exploiting a property called submodularity. Simply put, submodularity means there is a diminishing return associated with adding sensors — adding a sensor to a five-sensor network has much more impact than adding a sensor to a 10,000-sensor network. The algorithm also takes into account the property of locality — the idea that sensors that are far apart provide almost independent information.

Watch more breaking news now on our video feed:

Bookmark http://universeeverything.blogspot.com/ and drop back in sometime.

Video: U.S. Outsourcing Emissions, Too, Resarchers Say

Increased trade with other nations over the last decade, including China, has caused the United States to outsource more of its greenhouse gas emissions, according to two Carnegie Mellon University researchers.

"We've seen that although our consumption level has gone up considerably over the last decade, our greenhouse gas emissions have not gone up accordingly, which is mostly due to the fact that these emissions have been outsourced to the other countries who are making all the goods we are consuming here," says Christopher Weber, Carnegie Mellon researcher, Civil & Environmental Engineering.

The percentage of emissions from imports has nearly doubled over about the last decade, says H. Scott Matthews, associate rofessor, Civil & Environmental Engineering/Engineering & Public Policy.

Emissions from international trade, therefore, need to be better taken into account in global negotiations over climate, they say.

Watch a quick, five-minute video clip of Weber and Matthews describing their work on demand on Energy Policy TV. Visit the Climate Change Channel and click on "Trade Emissions."

Bookmark http://universeeverything.blogspot.com/ and drop back in sometime.

Online Shoppers Will Pay Extra to Protect Privacy, Carnegie Mellon Study Shows

People are willing to pay extra to buy items from online retailers when they can easily ascertain how retailers’ policies will protect their privacy, a new Carnegie Mellon University study shows.

Participants in the laboratory study used a Carnegie Mellon shopping search engine called Privacy Finder, which can automatically evaluate a Web site’s privacy policies and display the results on the search results page. The study, led by Lorrie Cranor, director of the Carnegie Mellon Usable Privacy and Security (CUPS) Lab, found that people were more likely to buy from online merchants with good privacy policies, which were identified by Privacy Finder.

They were also willing to pay about 60 cents extra on a $15 purchase when buying from a site with a privacy policy they liked.

Findings from the study, the first to suggest that people will pay a premium to protect their privacy when shopping online, will be presented Friday, June 8, at the Workshop on the Economics of Information Security, an international meeting hosted by Carnegie Mellon that begins Thursday, June 7. In addition to Cranor, the research team included Alessandro Acquisti, assistant professor of information technology and public policy, and graduate students Janice Tsai and Serge Egelman.

Many people express concerns that unscrupulous online merchants might misuse credit information, target spam to their email addresses or otherwise violate their privacy. But a number of previous studies have found that many people still fail to act to protect their privacy online. Some have shown that people willingly give up private information in return for lower prices or even the mere chance of a monetary reward.

“Our suspicion was that people care about their privacy, but that it’s often difficult for them to get information about a Web site’s privacy policies,” says Cranor, an associate research professor of computer science and of engineering and public policy. A Web site’s policies may not be readily accessible, can be hard to interpret and sometimes are nonexistent, Cranor says. “People can’t act on information that they don’t have or can’t understand,” she adds.

Privacy Finder is a search engine developed by Cranor and her students to address this issue. The engine makes use of the Platform for Privacy Preferences (P3P), a technical standard for creating machine-readable privacy policies. About 10 percent of Web sites overall and more than 20 percent of e-commerce sites now employ P3P, Cranor says, and of the top 100 most-visited Web sites, about a third use P3P.

The search engine can automatically read and evaluate the policies of Web sites that employ P3P, and it displays this information as a series of colored squares that indicate to the user whether the site’s privacy policy complies with his or her privacy preferences.


Bookmark http://universeeverything.blogspot.com/ and drop back in sometime.

Carnegie Mellon Professors Question Advice for Nuclear Attacks

In the current Fox television adventure series, 24, a terrorist explodes a small nuclear bomb in Los Angeles. In the May issue of the journal Health Physics, Carnegie Mellon researchers Keith Florig and Baruch Fischhoff offer simple, practical advice that ordinary citizens can use when faced by such threats.

Specifically, the two scientists address the following questions: whether it is worth citizens' time to stock supplies needed for a home shelter, how urgently should one seek shelter following a nearby nuclear detonation, and how long should survivors remain in a shelter after the radioactive dust settles.

Carnegie Mellon's Florig, a senior research engineer and Fischhoff, a university professor, report that many families simply can't afford the government stocking guidelines; they need help to protect themselves.

"A number of emergency-management organizations recommend that people stock their homes with a couple dozen categories of emergency supplies," says Florig of Carnegie Mellon's engineering and public policy department. "We calculated that it would cost about $240 per year for a typical family to maintain such a stock, including the value of storage space and the time needed to tend to it."

Their research also suggests that many families who could afford to follow the stocking guidelines might think twice about whether the investment was really worth it, given the low probability that stocked supplies would actually be used in a nuclear emergency.

"Government websites such as Ready.gov recommend that people take shelter or evacuate following a nuclear blast, but provide no information that might help people determine how much time they have to react before a fallout cloud arrives," says Florig. We advocate a more nuanced message with simple rules for minimizing risk based on how far people are from the blast. If you are within several miles of the blast, there will be no time to flee and you will have only minutes to seek shelter. If you are 10 miles from the blast, you will have 15 to 60 minutes to find shelter, but not enough time to reliably flee the area before the fallout arrives."

Finally, the researchers analyze how long people should remain sheltered in a contaminated area before it is riskier to stay than to evacuate.

"The answer depends on how good their shelters are and how long it would take to evacuate. Those who have poor shelters, limited stores and no access to a vehicle will need the most help to escape," according to the researchers.

Understanding these simple rules can help people to plan for themselves and help officials to plan for them.

Bookmark http://universeeverything.blogspot.com/ and drop back in sometime.