From: Real-time spatio-temporal event detection on geotagged social media
\(\Gamma _j\) | Region j where \(\Gamma =\bigcup \Gamma _j\) contains all posts in \({\mathcal {S}}\) |
\({\mathcal {W}}_{ij}\) | The set of posts in the sliding window at interval i for region j |
\(C_{ij}\) | Number of posts in the window slide increment interval i for region j |
\(\lambda _{ij}\) | Estimate rate of posts at interval i for region j |
\(P_{ij}\) | Poisson signal at interval i for region j |
\(\tau _1\) | Poisson signal threshold |
\(F_{ij}\) | Event signal at interval i for region j |
\(\tau _2\) | Event detection threshold |
\(\alpha\) | Event signal decay parameter |
\({\theta }_{duration}\) | Event duration threshold |
\({\theta }_{entity}\) | Minimum entities threshold |
\({\theta }_{area}\) | Quad-tree node region area threshold |
\({\theta }_{count}\) | Quad-tree node post count threshold |