Australian IT JOBS : Sydney IT jobs, UNIX jobs, Linux jobs, Java jobs, ASP jobs Linux.conf.au Linux.conf.au
Technology news and Jobs arrow Telecommunications arrow NEC's VoIP spam detector turns Turing Test on its head
NEC's VoIP spam detector turns Turing Test on its head PDF Print E-mail
Written by Stuart Corner   
Saturday, 27 January 2007
A computer can pass the Turing Test if it can fool a human being - through its responses in a conversation - into thinking that it is human. NEC's VoIP spam detector claims to be using a computer to determine whether a caller is human or a machine. Quite an achievement!

This is especially so since, according to Wikipedia, no computer has yet passed the Turing Test, which was devised by computing genius Alan Turing in 1950.

NEC will debut its new SPIT (spam over Internet telephony) blocker, called VoIP SEAL, at the 3GSM congress in Barcelona in February. The company has given few details of its workings other than to say that it uses a three stage approach to detect and block SPIT calls.

- Calls arising from spam-generating-software and calls from real individuals are separated by a Turing test. Before connecting the call, VoIP SEAL detects and blocks the unauthorised access based on the communication pattern observed during a call.

- By adopting a module structure, VoIP SEAL enables rapid response to new kinds of SPIT attacks, without adjusting the system, by adding and updating modules to respond to new and different kinds of SPIT.

- The adoption of a module structure also realises response to a broad range of applications by enabling flexible and easy customisation of systems to meet the needs of a variety of hardware, such as SIP servers, SBC), home network equipment and terminal equipment.

NEC claims that, in a simulation 99 percent of SPIT was detected and blocked, preventing users from receiving unwanted and bothersome calls. The company has given no indication of when it wil launch a commercial product.

* According to Wikipedia, "Trying to pass the Turing test in its full generality is not, as of 2005, an active focus of much mainstream academic or commercial effort. Current research in AI-related fields is aimed at more modest and specific goals. There is an ongoing $US10,000 bet at the Long Bet Project between Mitch Kapor and Ray Kurzweil about the question whether a computer will pass a Turing Test by the year 2029."{moscomment}


Get stories like this delivered daily - FREE - subscribe now
When you subscribe get a 12 months license for LiveProject
Valued at $99 USD


LiveWire - Desktop alerts Download the FREE iTWire desktop alert widget LiveWire - Desktop alerts


Del.icio.us!
 
< Prev   Next >
Contact , Register , Advertise with iTWire , Links , Register , About iTWire , Feedback , Post your jobs , Events , iTWire site map , Start Blogging
Industry Releases , Submit your release now , Start submitting to iTWire , How to post video