Ten years ago, more or less, Internet users were faced with a major problem with no obvious solution: spam. At the time, over half of all email traffic was thought to be spam, and that number was rising. Most people spent a significant amount of time every day weeding through junk mail, and it was considered a serious enough problem that a federal law (the CAN-SPAM Act of 2003) was passed to try to curb it. The law was not especially successful and today over 75 percent of email traffic is reportedly spam, but most people don’t see nearly as much of it anymore. What changed?

The answer is that, in around 2003, spam filtering software got much, much better due in large part to a change in philosophy. Until then, most filtering software had relied on keyword searches to identify spam, but a series of papers in 2002 and 2003 identified methods by which software could be “trained” to identify spam based on user input. Given a group of messages identified by the user as either “spam” or “not-spam,” these new filters could rapidly and accurately learn to identify unwanted email. This method turned out to be far more accurate than searching for human-generated keywords, and its accuracy could be improved over time with input from the user about any classification errors. Importantly, the new method was extremely effective at avoiding “false positives”—that is, messages that were actually not spam being marked as spam. Generally users would rather read through a few spam messages than miss an important non-spam email.

This content has been archived. It is available through our partners, LexisNexis® and Bloomberg Law.

To view this content, please continue to their sites.

Not a Lexis Subscriber?
Subscribe Now

Not a Bloomberg Law Subscriber?
Subscribe Now

Why am I seeing this?

LexisNexis® and Bloomberg Law are third party online distributors of the broad collection of current and archived versions of ALM's legal news publications. LexisNexis® and Bloomberg Law customers are able to access and use ALM's content, including content from the National Law Journal, The American Lawyer, Legaltech News, The New York Law Journal, and Corporate Counsel, as well as other sources of legal information.

For questions call 1-877-256-2472 or contact us at [email protected]