Live Search is researching a ranking feature similar to Google’s Page Rank called Browse Rank.

After being jilted by a lover , is it natural to seek solace in the arms of another?   Please allow us this brief moment to be dramatic since we are going to press on to a little-known document, to Google fans anyway, that was a tedious read.

It’s no secret that Microsoft is itching to take over Google’s place with Live Search. A gargantuan ambition worthy of the software giant considering that it places a distant third in the U.S. search market behind Yahoo and Google. Data from ComScore shows that 8.3% of searchers  use Microsoft’s search engine while 63% uses Google.

Looks like Microsoft has another trick up it sleeves apart from that Yahoo take-over bid. Researchers were busy drawing up a ranking scheme that will rival that of Google’s Page Rank. However, instead of an easily gamed  system relying on incoming links, Microsoft mainly uses visitors’ browsing behavior gathered from browsers and servers. It is not yet clear though how they will implement the infamous little green pixels.

Microsoft’s method is interesting, especially the way they disclosed their algorithm. Google’s search and ranking algorithm is a secret and it has spawned a whole web industry of guesswork and gaming.

We are quoting the abstract with words in bold characters for emphasis by us below.

This paper proposes a new method for computing page importance referred to as BrowseRank. The conventional approach to compute page importance is to exploit the link graph of the web and to build a model based on that graph. For instance, PageRank is such an algorithm, which employs a discrete-time Markov process as the model. Unfortunately, the link graph might be incomplete and inaccurate with respect to data for determining page importance, because links can be easily added and deleted by web content creators.

In this paper, we propose computing page importance by using a ’user browsing graph’ created from user behavior data. In this graph, vertices represent pages and directed edges represent transitions between pages in the users’ web browsing history. Furthermore, the lengths of staying time spent on the pages by users are also included. The user browsing graph is more reliable than the link graph for inferring page importance.

This paper further proposes using the continuous-time Markov process on the user browsing graph as a model and computing the stationary probability distribution of the process as page importance. An efficient algorithm for this computation has also been devised. In this way, we can leverage hundreds of millions of users’ implicit voting on page importance. Experimental results show that BrowseRank indeed outperforms the baseline methods such as PageRank and TrustRank in several tasks.

The entire document can be downloaded here. It was released midyear of 2008.