Is Twitter Reputation the solution for Google Real Time Spam?
Google Real Time Search has shown a worrying vulnerability to Twitter spam that highlights the need for more effective filters on real time results.
Spam in Twitter Search results is a common thing, and it will last until algorithms will be able to evaluate spam intent behind the use of a particular word or a hashtag in a tweet.
One debatable way of avoiding spam results in Twitter Search would be adding an authority filter option to results, that could rank tweets according to different factors. Twitalyzer is a tool that evaluates Twitter users across different dimensions: Influence, signal-to-noise ratio, generosity, velocity and many others.
Influence is calculated on factors that could be used to solve the real time spam filtering problem:
- Popularity: the number of followers you have (but as Robert Scoble points out, this metric can be easily gamed)
- Authority: measured by the number of times you are retweeted (the retweet button could make this trust metric easily measurable)
- Generosity: the number of times you retweet others (is it a good thing?)
- Clout: the number of times you are referenced by others
- Velocity: the number of updates published in a seven day period
Real Time Search Ranking Factors
Other factors like the number of times a tweet is favorited (and by whom it is favorited?), or link CTR could be important in measuring the reputation of a source. Retweet/follower or favorited/follower ratios could be part of the formula: The smaller the number of followers, the greater the effect favorited tweets and retweets would have in real time rankings.
Italian blogger Tagliaerbe understands that popularity isn’t an accurate measure of relevancy in Twitter and expands this concept introducing a list-related factor, that adds social recognition to topical relevancy: Inclusion in Twitter lists named with relevant terms would make a source more trustworthy and authoritative on a specific subject.
A Twitter Authority or Reputation Score could be the 201st ranking factor in Google’s algorithm. My optimistic guess is that it’s already in the development pipeline, or maybe it’s already there but it just needs some more tweaking.
Image credit: swanthinks.wordpress.com
0
Loading ...