Skip to main content

Table 4 Differences in topic distributions in terms of Jensen–Shannon distance (\(JS\))

From: Evolution of topics and hate speech in retweet network communities

  Timepoint t Community
Community 0 22 68 91 132 Left Right SDS Sports
Left 0.052 0.051 0.060 0.008 0.057 0.0 0.146 0.172 0.406
Right 0.051 0.047 0.049 0.019 0.034 0.0 0.092 0.481
SDS 0.101 0.114 0.091 0.028 0.044 0.0 0.482
Sports 0.074 0.020 0.036 0.087 0.082 0.0
  1. The left-hand side of the table shows the \(JS\) distances for each community C, between its cumulative distribution \(C_{\cup }\) and individual timepoints \(C_t\), \(JS (C_{t} \parallel C_{\cup })\). The right-hand side is a symmetrical matrix, with the \(JS\) distances between the cumulative distributions for all pairs ij of communities, \(JS (Ci_{\cup } \parallel Cj_{\cup })\). In bold are the \(JS\) distances \(0.1 < JS \le 0.4\), and in italics \(0.4 < JS\)