imagesThe ever-increasing amount of information flowing through social media forces the members of these networks to compete for attention and influence by relying on other people to spread their message. How is it that certain topics manage to get more attention than others, thus going viral and changing the agenda of an online community? A recent study published by HP Labs suggests that most Twitter users are generally rather passive and that the Twitter users with the most influence on the global stream are those who get off their thumbs and retweet.

Dr. Bernardo A. Huberman, the director of HP Labs’ Social Computing Lab, oversaw this research on the nature of user influence on social media networks such as Twitter. After analyzing 22 million tweets, Dr. Huberman and his co-authors calculated a novel measure of influence for individual users and developed a corresponding algorithm that automatically identifies particularly influential users. The algorithm described in the research is unique in that it incorporates what the authors call “passivity.”

The study found that a large majority of Twitter users act as passive information consumers and rarely forward (“retweet”) content to the network. To become influential, users must not only catch the attention of their followers; they must also overcome their followers’ predisposition to remain passive.

According to the research, it is important to separate the concept of “influence” from “popularity.” While a user on Twitter may have a large number of followers, his or her influence is more strongly associated with their engagement with the network, rather than the raw number of followers or retweets. To identify Twitter influencers, the authors devised an algorithm called the IP Algorithm. This algorithm assigns a relative influence score and passivity score to every user:

  • “Influence” depends on both the quantity and quality of the user’s audience
  • “Passivity” is a measure of how difficult it is for other users to influence him or her

In general, the model makes the following assumptions

  • A user’s influence score depends on the number of people he influences as well as their passivity.
  • A user’s influence score depends on how dedicated the people he influences are. Dedication is measured by the amount of attention a user pays to a given one as compared to everyone else.
  • A user’s passivity score depends on the influence of those who he’s exposed to but not influenced by.
  • A user’s passivity score depends on how much he rejects other user’s influence compared to everyone else.

An evaluation performed with a 2.5 million user dataset shows that the influence measure is a good predictor of URL clicks, outperforming several other measures that do not explicitly take user passivity into account. Interestingly, because robot accounts (which automatically aggregate keywords or specific content from any user on the network) “attend” to all existing tweets and only retweet certain ones, the percentage of information they forward from other users is actually very small and explains why the IP-algorithm assigns them high passivity scores.

The paper concludes: “This study shows that the correlation between popularity and influence is weaker than it might be expected. This is a reflection of the fact that for information to propagate in a network, individuals need to forward it to the other members, thus having to actively engage rather than passively read it and cease to act on it.” The authors also see their measure of influence applicable to other social networks.

With advertising and marketing initiatives increasingly relying on social media for their success, we will watch closely as the science of measuring influence matures, fed by research such as this.