Kim Kap-soo: raw computed results
The raw ranking from the production similarity calculation (face-embedding centroid cosine → gender-cohort z-normalization). The web-opinion crown, curation, and fusion re-rank are removed.
- 1
89.5%z 4.80 - 2
89.1%z 4.76 - 3
88.4%z 4.70 - 4
87.1%z 4.60 - 5
85.7%z 4.48 - 6
85.2%z 4.44 - 7
85%z 4.42 - 8
83.3%z 4.29 - 9
82.9%z 4.26 - 10
82.5%z 4.22 - 11
81.9%z 4.17 - 12
79.4%z 4.01 - 13
77.5%z 3.99 - 14
77.3%z 3.99 - 15
76.4%z 3.98 - 16
71.6%z 3.92 - 17
70.6%z 3.91 - 18
70.2%z 3.91 - 19
66.4%z 3.87 - 20
66.3%z 3.87 - 21
65.7%z 3.86 - 22
62.5%z 3.83 - 23
60.6%z 3.80 - 24
59.6%z 3.79 - 25
59.6%z 3.79 - 26
59.3%z 3.79 - 27
55%z 3.74 - 28
54.7%z 3.74 - 29
54.5%z 3.74 - 30
52.7%z 3.72 - 31
52.3%z 3.71 - 32
52.2%z 3.71 - 33
51.5%z 3.71 - 34
51.4%z 3.70 - 35
49.8%z 3.69 - 36
48.5%z 3.67 - 37
47.5%z 3.64 - 38
47.4%z 3.63 - 39
47.2%z 3.62 - 40
47.2%z 3.62 - 41
47.2%z 3.61 - 42
46.9%z 3.60 - 43
46.9%z 3.60 - 44
46.7%z 3.58 - 45
46.2%z 3.55 - 46
46.1%z 3.54 - 47
46%z 3.54 - 48
45.6%z 3.51 - 49
45.2%z 3.49 - 50
45.1%z 3.48