maandag 27 februari 2012

Measuring User Influence in Twitter: The Million Follower Fallacy

Meeyoung Cha, Hamed Haddadi, Fabricio Benevenuto, Krishna P. Gummadi, 2010. AAAI Publications, Fourth International AAAI Conference on Weblogs and Social Media, 10-17

There have already been previous study on influence in different domains, for example in the domain of marketing. In this paper, the authors focused their study on the social media Twitter. They wanted to know what kind of influence twitter users can have on each other.
3 Measures of influence:
In their research they make a contrast between 3 measures of influence. The first one is the number of followers people have on twitter (called indegree). Their conclusion is that this measure shows the popularity of the user, but it doesn't show the influence on other users. The second one is the number of retweets a user has, here they saw that the content value of a tweet has an important role . The last one is the number of mentions a user gets, this depends of the value of a name.
Connection between indegree, retweets and mentions:
Next, they studied the effect of the 3 measures on each other. They found that there is a link between the number of retweets and mentions. Furthermore they concluded that indegree has no relation with the other two measures. The last item they investigated was if the influence of a user changed over time. The result of their investigation is that users have to do efforts to maintain their influence.

Bryan Declerck

2 opmerkingen:

  1. After I’ve read your summary of “Measuring User Influence in Twitter: The Million Follower Fallacy”, I noticed that there is a link between your article and “Influence and passivity in social media”, the article I have read.
    Both articles talk about the influence social media users, especially Twitter users, can have on each other. However, the way the authors of “Influence and passivity in social media” operate is different from the other research. They work with the concept of passivity in a social network, which separates their method to measure influence from previous ones. I think it’s interesting to compare both articles, particularly their results.

    Mark Cazander

  2. I think it’s an interesting idea to study how people are affected by other users of social media. This research is similar to the research in one of my articles, “twitter mood predicts the stock market”. This kind of studies are very important to businesses and therefore also to the economy. Because in the future, social media such as twitter will become even more important to companies than on this moment. Companies will probably invest more money on marketing at social media, due to the massive reach of these media. I think that this is a good article and that it’s worth looking into.

    Nino De Moor