Expert communication on Twitter: Comparing economists' and scientists' social networks, topics and communicative styles.

Experts increasingly use social media to communicate with the wider public, prompted by the need to demonstrate impact and public engagement. While previous research on the use of social media by experts focused on single topics and performed sentiment analysis, we propose to extend the scope by investigating experts' networks, topics and communicative styles. We perform social and semantic network as well language analysis of top tweeting scientists and economists. We find that economists tweet less, mention fewer people and have fewer Twitter conversations with members of the public than scientists. Scientists use a more informal and involved style and engage wider audiences through multimedia contents, while economists use more jargon, and tend to favour traditional written media. The results point to differences in experts' communicative practices online, and we propose that disciplinary ways of 'talking' may pose obstacles to an effective public communication of expert knowledge.

Memory and burstiness in dynamic networks.

A discrete-time random process is described, which can generate bursty sequences of events. A Bernoulli process, where the probability of an event occurring at time t is given by a fixed probability x, is modified to include a memory effect where the event probability is increased proportionally to the number of events that occurred within a given amount of time preceding t. For small values of x the interevent time distribution follows a power law with exponent -2-x. We consider a dynamic network where each node forms, and breaks connections according to this process. The value of x for each node depends on the fitness distribution, ρ(x), from which it is drawn; we find exact solutions for the expectation of the degree distribution for a variety of possible fitness distributions, and for both cases where the memory effect either is, or is not present. This work can potentially lead to methods to uncover hidden fitness distributions from fast changing, temporal network data, such as online social communications and fMRI scans.