The Hindu Business Line : The company you keep

EVERYONE has heard of them. Everyone who checks e-mails encounters them. But do they know that the original SPAM is a brand of processed meat or "Spiced Ham"? Even Nikita Khrushchev credited SPAM with sustaining the Soviet Army during World War II. He said in his memoirs: "Without SPAM, we would not have been able to feed our army."

An anonymous bard versified with "gluttonous glee" his taste for this 70-year-old delicacy, little knowing his words would come true in a virtual world:

Long have my arteries clogged to the sound

of sizzling SPAM when there's no one around

furtively chewing or swallowing whole.

Triple bypass by forty, my medical goal.

Spam or Self Promotional Advertising Message is acknowledged to be the cholesterol of the Internet which "clogs everything". Even sanguine surfers agree that war against Spam has never looked bleaker, as about 60 per cent of all e-mails are Spam.

At the World Economic Forum in Davos, Switzerland, in January, the Microsoft Chairman, Mr Bill Gates, predicted that Spam would soon be a thing of the past, thanks to software being developed at his company. But beleaguered computer users are unlikely to believe this until they see it.

Anti-spam programs mostly work by scanning e-mail messages for the keywords that spammers use, which genuine friends tend to avoid. The technique goes by the formidable name of Bayesian Filtering and uses probability to work out if a mail is junk or real. It works in 99.7 per cent of the cases. But the spammers are clever, they soon learn the keywords under surveillance and avoid or disguise them. In this war between mail-spammers and mail-washers hope springs eternal in the hearts of e-mail addicts. Off and on there is news of one more algorithm to combat spam. Remember, when we were young and forming early friendships, our mothers told us that we are known by the company we keep. They even warned us becoming guilty by association.

This very principle of identifying a mutually friendly set distinct from others is the key to the newly successful anti-spam filter devised by Boykin and Roychoudhari of the University of California at Los Angeles as reported in a recent science magazine.

Their technique exploits the structure of social networks to quickly determine whether a given message comes from a friend or a spammer. It never rejects legitimate messages under the impression that they are spam.

Most people's e-mail comes from a limited social network, and these networks tend to be clustered into clumps where everyone knows each other. If Ramu knows and e-mails Kittu and Bablu, for example, then Kittu and Bablu are far more likely to know and e-mail each other than if they did not share a friend in common.

E-mails radiating from a spam source do not share this clustering property — the vast majority of recipients do not know each other. Quantifying the clustering of incoming e-mails, and eliminating them does the trick.

In any case, if you wish to avoid spam totally it is better to be alone, as George Washington said in his Rules of Civility.

More Stories on : E-Mail | Viruses | Impressions

Article E-Mail :: Comment :: Syndication :: Printer Friendly Page

Stories in this Section
A rush of new standards

BPO: Beyond information technology

BPO fallout: Youth shelving books for bucks

Air turbulence

Right firepower for economy

An advisable alternative

Debit ignorance and credit knowledge

Sun shines and wind blows but where's the auditor?

The company you keep

Rising prices