The performance for all the simulated neurons (each represented by a symbol with a unique combination of color and shape) is shown for each algorithm in Figure \ref{821904}. The accuracy of Spikes_Link_WC was significantly larger than the one of the other algorithms (p<2e-19). This difference is produced mainly by the lower recall obtained for the other algorithms (p<1e-15). A lower recall value is expected when the algorithms do overclustering. On the other hand for some simulated neurons, the performance showed a lower precision, as different neurons were merged into the same class. Still, this was resolved better by Spikes_Link_WC , showing a significant improvement in precision against the other algorithms (p<2e-3) .