MSN Search claims to freeze out web spam

Microsoft is incorporating a new filtering technology into its forthcoming MSN Search technology, aiming to offer results clear of web spam.

The company unveiled a research project at its Silicon Valley campus in Mountain View which uses statistical analysis to locate spam web pages.

A spam web page is designed solely to fool search engines by linking keywords to web pages that the spammer wants to show up high in the search results.

Spammers are increasingly trying to weasel their way into search engine results, and Microsoft hopes that filtering them out can be one area where its tool can outshine Google's.

"This will be part of the [upcoming] Microsoft search," researcher Marc Najork told vnunet.com.

Microsoft has so far given little detail about its MSN Search project, which is designed to compete head on with Google. The new search technology is scheduled to be available this year.

A spokeswoman for Google said that the company has a large team of engineers working on the quality of its search index, but declined to comment on any specific actions it is taking to clean up search results.

In a sample of one billion web pages that Microsoft tracked for the research project, the company found that eight per cent were spam.

In one case, the Microsoft researchers claim to have found a webpage in Germany that would constantly create pages filled with pieces of text that were copied from random web pages, linked to a porn site.

This not only leads visitors away from the webpage from which the text was copied, but gives the spammer an indefinite source of keywords and key phrases to link to.

The researchers found the site after their analysis identified that German web pages on average were updated more often than those in other geographies.