Thursday, January 14

Ranking Real-Time Results Interview with Amit Singhal

Yesterday, Technology Review posted an interview with Amit Singhal, on How Google Ranks Tweets. According to Amit, one key is to find "reputed followers",
"You earn reputation, and then you give reputation. If lots of people follow you, and then you follow someone--then even though this [new person] does not have lots of followers," his tweet is deemed valuable because his followers are themselves followed widely...
It seems like pretty straightforward translation of PageRank with "following" as a form of link endorsement. In this vein, see also Daniel's TunkRank.

The interview goes on to mentions the use of geolocation in tweets as a next likely step. Amit also rightly points out that blogs and news organizations are important components of "real-time" search; it's not just tweets.


  1. Interesting -- though, I follow several companies and high profile companies on Twitter. As soon as I follow them most of them automatically follow me back. So unless they have QC that says "unfollow bots and spammers after they out themselves," it sounds like naively, spammers would just have to look like a legitimate account long enough for companies to re-follow them and never look back. Then, spam away as a trusted source!

  2. That sounds like the same problem as reciprocal linking on web pages. Those types of patterns can be detected and removed/addressed.

    I imagine that many well-known spam detection algorithms would naturally translate.