Skip to main content

Table 1 The effect of the choice of features on spectral clustering

From: Semi-supervised clustering of fractionated electrograms for electroanatomical atrial mapping

Optimal feature set \(\mu _1\) \(\mu _2\) \(\mu _{sc}\)
\(\xi _2\)         0.459 0.225 −0.234
\(\xi _2\) \(\xi _8\)        0.514 0.197 −0.317
\(\xi _2\) \(\xi _8\) \(\xi _7\)       0.491 0.205 −0.286
\(\xi _2\) \(\xi _8\) \(\xi _7\) \(\xi _5\)      0.521 0.193 −0.327*
\(\xi _2\) \(\xi _8\) \(\xi _7\) \(\xi _5\) \(\xi _1\)     0.495 0.206 −0.286
\(\xi _2\) \(\xi _8\) \(\xi _7\) \(\xi _5\) \(\xi _1\) \(\xi _4\)    0.492 0.235 −0.257
\(\xi _2\) \(\xi _8\) \(\xi _7\) \(\xi _5\) \(\xi _1\) \(\xi _4\) \(\xi _3\)   0.483 0.235 −0.248
\(\xi _2\) \(\xi _8\) \(\xi _7\) \(\xi _5\) \(\xi _1\) \(\xi _4\) \(\xi _3\) \(\xi _6\) 0.450 0.243 −0.207
  1. Notation (\(^\mathbf{* }\)) points out on the selected feature subset, \({\varXi }^{*},\) that reaches the lowest value of \(\mu _{sc}\)