Skip to main content
Fig. 10 | Computational Social Networks

Fig. 10

From: Tracking online topics over time: understanding dynamic hashtag communities

Fig. 10

Statistics of various observables: a the correlation between the average likes \(\langle L_i/S_i \rangle\) per hashtag of community i and its size \(S_i.\) b The distribution of (community-)sizes in each snapshot S(t), plotted on top of each other and as a guide to the eye, a fitted power law (exponent: \(1.27\pm 0.03,\) KS-statistic: 0.2, p value: 0.08). c The distribution of relative gains \(\Delta S / S = (S(t)-S(t-1))/S(t-1)\) (green) in size and d relative losses \((\Delta S / S)_r = (S(t-1)-S(t))/S(t)\) (red). The Inset shows the distribution of interburst times \(\Delta t\) between events of \(\Delta S / S > 10\) (blue) and a fitted power law (exponent: \(1.1\pm 0.02,\) KS-statistic: 0.3, p value 0.05)

Back to article page
\