Ngram calculation results

Evaluation against standard

NComb.上/下ScoreComm.
n2Count3/12135
n2Comb.prop3/12126
n3Count3/12057
n3Comb.prop3/12056
n2Position3/11956
n3Position3/11877
n4Count3/11715
n4Comb.prop3/11716
n4Position3/11705
n5Count3/11586
n5Comb.prop3/11586
n5Position3/11506
n2Count25/151383
n2Comb.prop25/151383
n2Position25/151353
n2Position50/40794
n5Position25/15603
n3Position25/15593
n2Count50/40584
n2Comb.prop50/40584
n5Position50/404510
n3Count25/15423
n3Comb.prop25/15423
n5Count50/40429
n4Position25/15403
n5Count25/15373
n5Comb.prop25/15373
n3Count50/40265
n3Comb.prop50/40264
n4Count25/15263
n4Comb.prop25/15263
n5Comb.prop50/402610
n3Position50/40254
n4Count50/40184
n4Comb.prop50/40183
n4Position50/40174

Self(=same 部) / Other (=other 部) results

Results for different values of the upper and lower limits (上下) and vocabulary size (3万, 2万).

For all subcategories 部類 in a category 部, for the 10 most similar subcategories, a same/other calculation is performed, the maximal score is thus 10 times the number of the subcategories for a given category.

KR1 經部 (10部類)

Results for KR1
LimitCountProb.Pos
上下n2n3n4n5n2n3n4n5n2n3n4n5
50/4017.00 5.00 11.00 41.00 17.00 5.00 8.00 13.00 26.00 12.00 7.00 8.00
25/1548.00 22.00 10.00 7.00 48.00 22.00 10.00 7.00 55.00 30.00 14.00 9.00
3/161.00 53.00 51.00 48.00 60.00 53.00 51.00 48.00 58.00 51.00 50.00 47.00

KR2 史部 (16部類)

Results for KR2
LimitCountProb.Pos
上下n2n3n4n5n2n3n4n5n2n3n4n5
50/4025.62 25.62 42.50 49.38 25.62 25.62 40.00 30.00 27.50 18.75 28.75 15.62
25/1563.75 43.12 50.00 50.62 63.75 43.12 50.00 50.00 70.62 48.12 47.50 50.62
3/159.38 57.50 56.25 54.37 59.38 57.50 56.25 53.75 58.13 56.88 58.13 56.25

KR3 子部 (12部類)

Results for KR3
LimitCountProb.Pos
上下n2n3n4n5n2n3n4n5n2n3n4n5
50/4035.83 29.17 13.33 22.50 35.83 29.17 15.00 35.00 35.00 32.50 22.50 31.67
25/1529.17 30.00 23.33 25.00 29.17 30.00 23.33 25.83 26.67 31.67 26.67 27.50
3/132.50 29.17 30.00 32.50 31.67 30.00 30.00 32.50 32.50 30.00 30.00 30.83

KR4 集部 (9部類)

Results for KR4
LimitCountProb.Pos
上下n2n3n4n5n2n3n4n5n2n3n4n5
50/4044.44 45.56 31.11 4.44 44.44 45.56 34.44 22.22 44.44 45.56 42.22 41.11
25/1548.89 38.89 30.00 24.44 48.89 38.89 30.00 24.44 48.89 40.00 30.00 24.44
3/138.89 41.11 40.00 36.67 37.78 41.11 40.00 37.78 36.67 42.22 41.11 37.78

KR5 道部 (9部類)

Results for KR5
LimitCountProb.Pos
上下n2n3n4n5n2n3n4n5n2n3n4n5
50/4036.67 4.44 6.67 0.00 36.67 5.56 8.89 12.22 38.89 13.33 6.67 3.33
25/1562.22 37.78 26.67 13.33 62.22 37.78 26.67 14.44 68.89 42.22 35.56 16.67
3/153.33 60.00 64.44 63.33 53.33 61.11 64.44 63.33 52.22 60.00 63.33 62.22

KR6 佛部 (21部類)

Results for KR6
LimitCountProb.Pos
上下n2n3n4n5n2n3n4n5n2n3n4n5
50/4059.05 13.81 12.86 0.00 59.52 13.81 17.14 27.14 79.05 14.76 15.71 21.90
25/15100.00 99.52 95.71 95.24 100.00 99.52 95.71 95.24 100.00 100.00 97.14 95.24
3/193.81 94.76 96.67 96.67 93.81 94.76 96.67 96.67 93.33 94.76 96.19 97.14