Skip to main content
Fig. 3 | BMC Evolutionary Biology

Fig. 3

From: A new fast method for inferring multiple consensus trees using k-medoids

Fig. 3

Classification performances of the four versions of our k-medoids tree clustering algorithm in terms of ARI with respect to the number of tree leaves: a the case of 2 to 5 clusters and b the case of 6 to 10 clusters. The four tested versions of our algorithm were based on: 1) SH with RF (), 2) CH with RF (×), 3) SH with RF squared () and 4) CH with RF squared (). The coalescence rate parameter in the HybridSim program was fixed to 5 in this simulation. The presented results are the averages taken over all considered numbers of clusters

Back to article page