Social Search Patent for Facebook

Just browsing over at Search Engine Land and read an interesting article by Greg Sterling about Facebook’s patented ‘Social Search’. Based on the premise that search results are calculated algorithmically according to the number of clicks from members of social networks, the original patent (filed in 2004) stated:

Search results, including sponsored links and algorithmic search results, are generated in response to a query, and are marked based on frequency of clicks on the search results by members of social network who are within a predetermined degree of separation from the member who submitted the query. The markers are visual tags and comprise either a text string or an image.

So what factors would contribute to Facebook’s relevancy ranking methodology? Here’s a roundup:

• number and history of clicks
• degrees of connections within a network
• Possibly Facebook ‘likes’ (although because the patent was filed in 2004, Likes were not a part of the original patent)

Of course this differs somewhat from how ‘traditional’ ranking works in search engines such as Google, and historically factors such as linking, keyword relevancy and click through rates have played hugely important roles to determine the position of a site for a given query.

Resources

• Here’s a link to the original patent
• Article source: Facebook patents social search