Posts Tagged ‘Research’

Software fingers fake entries

In ARTIFICIAL INTELLIGENCE on July 31, 2011 at 10:16 pm
View of the Gables Great Hall, Cornell Univers...


Just as some models do not work well, some forms of Artificial Intelligence does seem to work very well indeed. And sadly, we, flesh and blood people lose when models fail, and we lose when AI does work, because in both instances, we become a function of the machine.


If you’re like most people, you give yourself high ratings when it comes to figuring out when someone’s trying to con you. Problem is, most people aren’t actually good at it–at least as far as detecting fake positive consumer reviews.

Fortunately, technology is poised to make up for this all-too-human failing. Cornell University researchers have developed software that they say can detect fake reviews (PDF). The researchers tested the system with reviews of Chicago hotels. They pooled 400 truthful reviews with 400 deceptive reviews produced for the study, then trained their software to spot the difference.

The software got it right about 90 percent of the time. This is a big improvement over the average person, who can detect fake reviews only about 50 percent of the time, according to the researchers.

They say people fall into two camps. One type accepts too much at face value and doesn’t reject enough fake reviews. The second type is overly skeptical and rejects too many real McCoys. Despite their very different approaches, each camp is right about half the time.

The Cornell system is similar to software that sniffs out plagiarism. While the plagiarism software learns to spot the type of language a specific author uses, the Cornell software learns to spot the type of language people use when they’re being deceptive in writing a review, said Myle Ott, the Cornell computer science graduate student who led the research.

The software showed that fake reviews are more like fiction than the real reviews they’re designed to emulate, according to the researchers. In part, deceptive writers used more verbs than real review writers did, while the real writers used more punctuation than the deceptive writers. The deceptive writers also focused more on family and activities while the real writers focused more on the hotels themselves.

The research team’s next steps are to use the technique with other types of service reviews, like restaurant reviews, and eventually try it with product reviews. The idea is to make it harder for unscrupulous sellers to spam review sites with fictitious happy customers.

Of course, just about any technology can be used for good or evil. The Cornell fake review spotter “could just as easily be used to train people to avoid the cues to deception that we learned,” Ott said.

This could lead to an arms race between fake review producers and fake review spotters. Ott and his colleagues are gearing up for it. “We’re considering… seeing if we can learn a new set of deception cues, based on fake reviews written by people trained to beat our original system,” he said.

Answer: the one on the right is fake.

Cornell software fingers fake online reviews | Crave – CNET.



Crowd-simulating software OR the ability to kill masses of people FAST

LDA entrance in Palestra House, designed by Wi...


GIST OF IT: While crowd simulation software has been developed before, the Bath/Bournemouth team hopes to use modern advances in processing power to create a more sophisticated program that models hundreds or thousands of individuals’ movements.

WHY SCYNET CARES:  Again, we feel our sense of identity being stripped away. Welcome to efficiency in the machine. This is the way prisons used to be constructed. Soulless,  while buildings used to be something that is both art and culture. Not anymore. Now its pure simulation. And beware of these studies that models movement. One day this very technology can of course find its way into  based upon models of their probable movement. Maybe the robot ship can fire at random spots where humans will be in the next fraction of time. Crazy fantastical rubbish? I hope so. In the meantime the technology will be constructing the space where YOU live.



A new project that uses artificial intelligence to model how crowds move could help architects design better buildings.  Researchers from Bath and Bournemouth universities are working with engineering consultancy Buro Happold to create software that shows how a building’s design can enable or prevent large numbers of people moving easily through it.

The program will create a visual representation of a crowd, modelling it as a group of many individual ‘agents’ instead of as a single mass of people and giving each agent its own goals and behaviour.

What Buro Happold wants to be able to understand is the impact of a space on the way people move,’ said Julian Padget, project supervisor and senior lecturer in computer science at Bath University.

‘There’s also the related question of what happens when a large volume of people are all trying to get somewhere rapidly, such as in an emergency situation.’

While crowd simulation software has been developed before, the Bath/Bournemouth team hopes to use modern advances in processing power to create a more sophisticated program that models hundreds or thousands of individuals’ movements.

The project will tackle the problems of simulating the crowds and rendering them in a believable way, from both a wide-angle and a close-up view, meaning the individuals have to appear realistic and show how their movements affect the rest of the group.

‘You don’t want it to look like a bunch of automatons wandering around — the reason being that it distracts the viewer, because they find it unnatural,’ said Padget. ‘They pay attention to that rather than what the picture overall is showing them.’

Instead of programming the computerised people with specific instructions, the computer will give them a destination and a range of actions to choose from and leave them to determine their own route, partly based on data gathered from observing real crowds.

But there are still limits to computational power and simulating greater numbers of people will require each individual character to have less intelligent programming, said Padget.

‘Our challenge is to work out what we can throw away from the sophisticated model and still get plausible-looking behaviour when we’ve got a large number of individuals.’

The simulation software will also need to be compatible with a suitable platform to render buildings designed by Buro Happold.

The four-year research project will be carried out by an engineering doctorate student through the universities’ Centre for Digital Entertainment, funded by the EPSRC.



Crowd-simulating software could improve building design | News | The Engineer.

Artificial Intelligence Diagnoses Abuse : Discovery News

There is no consensus on how closely the brain...

Image via Wikipedia

broken bone one day, a particular infection a few months later and depression the following year may appear to be separate, medical issues.

However, to a new artificial intelligence program developed by Boston doctors, these are all symptoms of domestic abuse.

The new software can identify abuse victims up to six years before these cases would otherwise be found and could eventually be used to diagnose just about any disease or injury.

“It’s very difficult to detect domestic abuse because it often happens in the privacy of the home,” said Ben Reis, a doctor at Children’s Hospital Boston (CHB) who helped develop the program.

“Doctors are often on the front lines of detecting abuse, but so often the doctor is focused on treating the injury, they don’t see the context behind it.”

Dozens of studies over the last 40 years have correlated various illnesses, injuries and other conditions with abuse. Bruising to the middle of the forearm or the core of the body instead of the elbow or knee can signal abuse. Depression or alcoholism may also be symptoms of this condition.