The Way Google Ranks Tweets – Official Google Statement by Amit Singhal
Posted on Tuesday, January 19, 2010
- Cloud Sites Update: New Control Panel & Faster MySQLThursday, April 9, 2015
- Introducing New Server Networking FeaturesTuesday, March 10, 2015
- Getting started with Docker in minutes using Docker MachineFriday, February 20, 2015
- MEAN & Socket.IO Integration TutorialThursday, September 11, 2014
- Getting Started with MEAN StackWednesday, August 6, 2014
- How To Install Ruby on Rails on UbuntuSunday, December 29, 2013
- Few CSS tricks you may not knowFriday, October 27, 2006
- How To Deploy Django on Nginx, Gunicorn with PostgresTuesday, April 8, 2014
- How to Add Swap on UbuntuMonday, December 30, 2013
- CentOS 6 to CentOS 7 Upgrade ProcedureWednesday, July 16, 2014
Google Search Engine started to show real-time results side by side to the regular search result pages. These real-time results are meant to show web searchers an access to new and instant news items as fast as it happen.
The main components of Google’s real-time results are twitter tweets. These are the real-time micro-blog messages that Twitter users use to post news and activites. Google’s Amit Singhal, who led the development of Google’s real-time search, recently showed how Google ranks these tweets in the new real-time resultson Google search pages.
There’s some kind of Page Rank only for these twitter tweets.
Google’s Page Rank algorithm for standard pages is to looks at the link structure of a webpage. The more links to a website and the more links to the linking websites show more relevant the linked website is.
But tweets from twitter are not about links but its about followers. People “follow” the comments or tweets of other Twitter users. The more followers a Twitter user has, again the more reputable the tweets are for that user. If Twitter users who have many followers follow another Twitter user then these users can have a bigger impact to the reputation of that user.
“It is more than a popularity contest”, said Google’s Amit Singhal. “One user following another in social media is analogous to one page linking to another on the Web. Both are a form of recommendation.Â As high-quality pages link to another page on the Web, the quality of the linked-to page goes up. Likewise, in social media, as established users follow another user, the quality of the followed user goes up as well.”
There are other filters and algorithms that rules this as the follower reputation rank is only one factor of Google’s methods to rank these tweets, other factors like hash tags, spam and the signal in the noise are some others.
Hash tags: Twitter users use “hash tags” in twitter comments. Hash tags are symbols (like keywords) that start with a # followed by a popular topic, as an example of hash tag #earthquake. If this hash tag is included in a tweet, this tweet will start to show up in the real-time results when other Twitter users click that hash tag’s topic word elsewhere on the site.
Spam: Hash tags can be very useful to maximize the exposure of a tweet, but sometimes they are abused for spamming. The wrong hash tags can become a red flag that triggers Google’s search spam filters. Amit Singhal didn’t go into the details in this but he said that Google modeled the hash tagging behavior in ways to reduce the exposure of spam or low-quality tweets.
The signal in the noise: There can be thousands of tweets that has a very welknown word like “Obama”. To find the best relevant tweets, Google searches for “signals in the noise”. Such signal can be a huge number of tweets that mention other words relative to “Obama”, for example “Cambridge police”. These kind of tweets with these kind of signals will be chosen for the real-time results.