Bing’s new three-column search interface has been kicking around the innernets for about a month now, integrating people’s Facebook and Twitter networks into queries so as to provide a social feature to search results. While it’s been interesting seeing what all my friends think they know about certain subjects, I’m not always certain I can take their word at face value (or maybe I don’t want to take their word at all). Luckily, there’s a second component to Bing’s social search that incorporates information from people who actually know a thing or two about a thing or two. Paul Yiu and the Bing Social Search Team gave a little more insight about how Bing populates its social results not only from your friends but from known trusted sources, too.
Essentially, Bing’s social search results are divided into two categories: Friends Who Might Know and People Who Know. While those categories could be vague synonyms from each other (assuming your friends are in fact people), the Friends Who Might Know category includes all of your direct contacts from your Facebook account who may have posted recently about your search topic or may even have it listed as one of their interests. The People Who Know is more affirmative for a reason: this section includes experts and enthusiasts who maintain some manner of influence or authority on your search topic. This will consist of results from Twitter and blogs.
For example, when I searched “New York City” on Bing, I got the following results:
While my single Facebook friend who likes New York City might be able to recommend a couple of places, it’s also helpful to have, say, the input of Gael Greene since she’s a pretty seasoned restaurant critic in the Big Apple. I also got a few reporters from the Times that write about NYC-related topics, so again, that could be interesting to follow-up on for more current goings-on in the city.
According to Yiu, People Who Know differs from it’s counterpart in that this category will be the same for all Bing users. After all, it wouldn’t make any sense to have different authorities for different users when the search is on the same subject, right? Since there is a lack of support from your immediate social network, Bing stacks up the experts according to the following metrics:
The way Bing identifies authorities on topics involves the search engine’s machine learning techniques that assess if and how people are popular for a specific topic and not just popular for the sake of itself. To highlight this fact, Yiu produced a rather satisfying example of this: “For example, Kim Kardashian is influential on Twitter but probably won’t appear for the query ‘machine learning.'” Hopefully, you never search for anything that gives Bing any reason to provide you with social search results from any Kardashian.
More about how to meaningfully use Bing’s social search results can be found on the Help forum for Bing.
[Via Bing Search Blog.]