Microsoft is using data-mining techniques from a third party in its internet browsers to guard against 'phishing' attacks, where dubious websites try to harvest personal information from unknowing victims.
Digital Resolve announced today that Microsoft is licensing data that comes from its Trusted Server technology, which crawls the internet and builds lists of websites and their legitimate IP addresses.
The data will be used by the antiphishing filter in Microsoft's Internet Explorer versions 6.0 and 7.0 in addition to Windows Live Toolbar, a browser search box, Digital Resolve said.
The technology focuses heavily on the websites of financial companies and e-commerce sites. It looks at 39 characteristics of a website, including IP address, location, domain and the ISP, said Dennis Maicon, Digital Resolve's vice-president for financial services solutions. It also checks how long a particular domain is valid based on its registration information.
"When somebody goes to type in Barclays.com, it's able to then take that URL, resolve it to an IP address, and then check against a list to see if it is an authorised Barclays server," Maicon said.
The company says its technology takes the opposite approach of so-called 'blacklists', which are lists of reported phishing sites, said Susan Daw, Digital Resolve's director of marketing. Keeping updated blacklists can be difficult as phishing sites may only be active for a short time, Daw said.
Rather than comparing a URL to a list of bad sites, Digital Resolve's technology lets the phishing filter know which sites are good, Daw said. Microsoft is using a combination of technologies for the phishing filter, including blacklists.
Daw said the technology is 100 percent successful for websites it has indexed. However, it's the first time the technology has been deployed, she said.
For suspicious websites, such as a bank site with an irregular URL, the phishing filter triggers a warning to the user to either continue viewing the site or close the web page.
Digital Resolve said the technology can guard against 'man in the middle'-style attacks, where an attacker collects a victim's login and password before passing the person to the legitimate site. Those attacks are often mounted from an irregular IP address that can be checked, according to Digital Resolve.