Calculates the element-wise element-centric similarity between two clustering results.
element_sim_elscore(
clustering1,
clustering2,
alpha = 0.9,
r_cl1 = 1,
rescale_path_type_cl1 = "max",
ppr_implementation_cl1 = "prpack",
dist_rescaled_cl1 = FALSE,
row_normalize_cl1 = TRUE,
r_cl2 = 1,
rescale_path_type_cl2 = "max",
ppr_implementation_cl2 = "prpack",
dist_rescaled_cl2 = FALSE,
row_normalize_cl2 = TRUE
)The first clustering result, which can be one of:
A numeric/character/factor vector of cluster labels for each element.
A samples x clusters matrix/Matrix::Matrix of nonzero membership values.
An hclust object.
The second clustering result, which can be one of:
A numeric/character/factor vector of cluster labels for each element.
A samples x clusters matrix/Matrix::Matrix of nonzero membership values.
An hclust object.
A numeric giving the personalized PageRank damping factor; 1 - alpha is the restart probability for the PPR random walk.
A numeric hierarchical scaling parameter for the first clustering.
A string; rescale the hierarchical height of the first clustering by:
"max" : the maximum path from the root.
"min" : the minimum path form the root.
"linkage" : use the linkage distances in the clustering.
Choose a implementation for personalized page-rank calculation for the first clustering:
"prpack": use PPR algorithms in igraph.
"power_iteration": use power_iteration method.
A logical: if TRUE, the linkage distances of the first clustering are linearly rescaled to be in-between 0 and 1.
Whether to normalize all rows in the first clustering so they sum to one before calculating ECS. It is recommended to set this to TRUE, which will lead to slightly different ECS values compared to clusim.
A numeric hierarchical scaling parameter for the second clustering.
A string; rescale the hierarchical height of the second clustering by:
"max" : the maximum path from the root.
"min" : the minimum path form the root.
"linkage" : use the linkage distances in the clustering.
Choose a implementation for personalized page-rank calculation for the second clustering:
"prpack": use PPR algorithms in igraph.
"power_iteration": use power_iteration method.
A logical: if TRUE, the linkage distances of the second clustering are linearly rescaled to be in-between 0 and 1.
Whether to normalize all rows in the second clustering so they sum to one before calculating ECS. It is recommended to set this to TRUE, which will lead to slightly different ECS values compared to clusim.
Vector of element-centric similarity between the two clusterings for each element.
Gates, A. J., Wood, I. B., Hetrick, W. P., & Ahn, Y. Y. (2019). Element-centric clustering comparison unifies overlaps and hierarchy. Scientific reports, 9(1), 1-13. https://doi.org/10.1038/s41598-019-44892-y
km.res = kmeans(iris[,1:4], centers=8)$cluster
hc.res = hclust(dist(iris[,1:4]))
element_sim_elscore(km.res, hc.res)
#> 1 2 3 4 5 6 7 8
#> 0.5725739 0.5883689 0.5713929 0.5646983 0.5953591 0.5775034 0.5888063 0.5725739
#> 9 10 11 12 13 14 15 16
#> 0.5793498 0.5883689 0.5552506 0.5678591 0.5967179 0.5884860 0.5526488 0.5526488
#> 17 18 19 20 21 22 23 24
#> 0.5522368 0.5725739 0.5775034 0.5618324 0.5643404 0.5618324 0.6104780 0.5594834
#> 25 26 27 28 29 30 31 32
#> 0.5678591 0.5967179 0.5594834 0.5857504 0.5857504 0.6113811 0.6113811 0.5643404
#> 33 34 35 36 37 38 39 40
#> 0.5403762 0.5403762 0.5883689 0.6163701 0.5642079 0.5953591 0.5793498 0.5725739
#> 41 42 43 44 45 46 47 48
#> 0.5784556 0.5631405 0.5793214 0.5677370 0.5808829 0.5883689 0.5705257 0.5646983
#> 49 50 51 52 53 54 55 56
#> 0.5552506 0.5784556 0.3971228 0.3888342 0.3971228 0.4145138 0.3990188 0.4431152
#> 57 58 59 60 61 62 63 64
#> 0.3888342 0.2642398 0.3990188 0.4160351 0.2068583 0.4492832 0.4076366 0.4053566
#> 65 66 67 68 69 70 71 72
#> 0.4009317 0.3942835 0.4431152 0.4669962 0.2400096 0.4045893 0.3563649 0.4492832
#> 73 74 75 76 77 78 79 80
#> 0.3621665 0.3966004 0.3946250 0.3942835 0.3971972 0.3841961 0.4063244 0.1490502
#> 81 82 83 84 85 86 87 88
#> 0.1722459 0.1722459 0.4655941 0.3621747 0.4431152 0.3834688 0.3968295 0.3005642
#> 89 90 91 92 93 94 95 96
#> 0.4737055 0.4145138 0.4431152 0.4053566 0.4655941 0.2642398 0.4695627 0.4688420
#> 97 98 99 100 101 102 103 104
#> 0.4688420 0.3946250 0.2405771 0.4695627 0.1746209 0.3670500 0.2355955 0.3720433
#> 105 106 107 108 109 110 111 112
#> 0.3720433 0.3045653 0.3606773 0.2427842 0.3370849 0.1678165 0.3612985 0.1778436
#> 113 114 115 116 117 118 119 120
#> 0.3652728 0.3674982 0.3412419 0.3559829 0.3747542 0.2934328 0.2828210 0.2494739
#> 121 122 123 124 125 126 127 128
#> 0.2448791 0.3571895 0.3045653 0.3573927 0.2277836 0.2587716 0.3573927 0.3571808
#> 129 130 131 132 133 134 135 136
#> 0.3747542 0.2587716 0.2427842 0.2934328 0.3747542 0.3621747 0.3160663 0.2744339
#> 137 138 139 140 141 142 143 144
#> 0.2797074 0.3747542 0.3571808 0.3652728 0.2274133 0.3652728 0.3670500 0.2448791
#> 145 146 147 148 149 150
#> 0.2274133 0.3652728 0.3311093 0.3612985 0.2797074 0.3411371