a) Visualization of the weighted networks (<i>W</i>^<i>t</i>) mapping evolution of communities over 5 time-stamps. <i>W</i>^<i>t</i> tracks evolution of a cluster between two consecutive time-stamps. The colors represent the weight of the edges in <i>W</i>^<i>t</i>. The weights can take value in the range [0, 1] i.e. 0 ≤ <i>w</i>(V^<i>t</i>(<i>j</i>, <i>k</i>)) ≤ 1. b) Visualization and tracking of community evolution by Netgram for the clusters obtained from the Kernel Spectral Clustering with Memory Effect (MKSC) algorithm [28] for Birthdeath dataset.

<p>Netgram showcases the birth, death, merge, split, expansion, shrinkage and continuation of communities for the Birthdeath dataset over 5 time-stamps (T₁, T₂, T₃, T₄ and T₅). We represent each community with a different colour circle and the size is ∝ the number of nodes in that community. From Fig 2b, we can observe the death of clusters <b>C6</b> & <b>C7</b> at time-stamps T₃ and T₄ respectively. Similarly, we can see birth of clusters <b>NewC14</b> & <b>NewC15</b> at time-stamp T₄ and T₅ respectively. We also observe that cluster <b>C11</b> merges with <b>C5</b> at time-stamp T₂ and cluster <b>C8</b> splits into 2 clusters at time-stamp T₄. We can observe that cluster <b>C8</b> has expanded at time-stamp T₃. The cluster <b>C8</b> also contracts at time-stamp T₄ as it splits into 2 clusters. Cluster <b>C1</b> demonstrates continuation over time.