Washington, July 16 (ANI): Computer scientists at Newcastle University have come up with a simple game that can turn a tedious manual labelling task into a form of light entertainment, and simultaneously help companies improve their chances of tackling online spammers.
Dr. Jeff Yan and his PhD student Su-Yang Yu call their innovation ‘Magic Bullet’.
The researchers highlight the fact that commercial websites like Google and Yahoo use Completely Automated Public Turing Test to Tell Computers and Humans Apart (CAPTCHA) to defend against malicious Internet bots, which spread junk emails or grab thousands of free email accounts.
They say that a common approach to testing its robustness is to try and attack or break the scheme.
According to them, this involves acquiring a set of labelled samples, but as computers find it difficult to recognise distorted test or images, this task usually falls to human researchers.
“Manually labelling samples is tedious and expensive. For the first time, this simple game turns it into a fun experience with a serious application as it also achieves a labelling accuracy of as high as 98 per cent,” says Dr. Yan.
Since spammers can misuse computer programs that can automatically bypass a heavily used CAPTCHA, it is important for researchers to understand and improve the robustness of the system in order to stay one step ahead.
To fully evaluate the robustness of a CAPTCHA scheme, at least 10,000 segments usually have to be labelled – a task that cannot be automated.
Dr. Yan and Yu say that their Magic Bullet is a dual-purpose online shooting game that can be played just for fun, but also contributes to solving a real problem.
Players are randomly pitched against each other, with two in each team. They cannot communicate with each other, and security techniques are used to ensure that they are geographically apart to reduce the likelihood of cheating.
Just in case there are not enough human players, one of two types of bots-a Data Relay Bot that replays data from old games or a Tailored Response Bot that acts according to an opposing team’s performance-will be introduced.
A randomly chosen segmented CAPTCHA character appears in each round, and shoots towards the target only when both players correctly identify it before their opponents.
Although the computer does not know which character each of the segments is, the answers given by the winning team can be accurate labels for the segments in the majority of cases.
The researchers have also included a high scoring table in the game in order to encourage players to return to improve on a previous score.
“An average game session produced 25 correct labels per minute, giving 1,500 per hour. Although this is not particularly fast, if touch typists were used it would be noticeably improved, and also players need time to get to know how the game works,” says Dr. Yan.
“As this game supports a large number of parallel sessions, which are limited only by the network bandwidth and game server’s CPU and memory, there is also a lot of scope to increase the labelling rate dramatically,” he adds.
A presentation on the research team’s findings were made at the ongoing IJCAI’09, a leading artificial intelligence conference in Pasadena, CA, USA. (ANI)