Online human interactions take place within a dynamic hi-erarchy, where social inuence is determined by qualities such as status,eloquence, trustworthiness, authority and persuasiveness. In this work,we consider topic-based Twitter interaction networks, and address thetask of identifying inuential players. Our motivation is the strong desireof many commerical entities to increase their social media presence byengaging positively with pivotal bloggers and tweeters. After discussingsome of the issues involved in extracting useful interaction data froma Twitter feed, we dene the concept of an active node subnetwork se-quence. This provides a time-dependent, topic-based, summary of rel-evant Twitter activity. For these types of transient interactions, it hasbeen argued that the ow of information, and hence the inuence of anode, is highly dependent on the timing of the links. Some nodes withrelatively small bandwidth may turn out to be key players because oftheir prescience and their ability to instigate follow-on network activity.To simulate a commercial application, we build an active node subnet-work sequence based on key words in the area of travel and holidays.We then compare a range of network centrality measures, including arecently proposed version that accounts for the arrow of time, with re-spect to their ability to rank important nodes in this dynamic setting.The centrality rankings use only connectivity information (who Tweetedwhom, when), but if we post-process the results by examining accountdetails, we nd that the time-respecting, dynamic, approach, which looksat the follow-on ow of information, is less likely to be `misled' by ac-counts that appear to generate large numbers of automatic Tweets withthe aim of pushing out web links. We then benchmark these algorith-mically derived rankings against independent feedback from ve socialmedia experts who judge Twitter accounts as part of their professionalduties. We nd that the dynamic centrality measures add value to theexpert view, and indeed can be hard to distinguish from an expert interms of who they place in the top ten. We also highlight areas wherethe algorithmic approach can be rened and improved.
|Publication status||Published - 5 Dec 2012|
|Event||SocInfo 2012 - Lausanne, Switzerland|
Duration: 5 Dec 2012 → 7 Dec 2012
|Period||5/12/12 → 7/12/12|
- dynamic targeting
- online social medium
- twitter interaction networks