Performance Cyberhate Classification
 Share
The version of the browser you are using is no longer supported. Please upgrade to a supported browser.Dismiss

 
View only
 
 
ABCDEFGHIJKLMNOPQRSTUVWXYZ
1
(UPSAMPLNIG N=3122)
2
Number Features
DOWNSAMPLING ( N=1622)
LogR1SVM1RF1
3
80173Gramsxxx
4
8072Gramsxxx
5
861embedded2xxx
6
10005embedded3xxx
7
50Doc2Vecxxx
8
50Word2Vecxxx
9
20Linguisticxxx
10
Accuracy0.65510.62060.60090.60590.64770.6650.56890.65270.62320.59850.63550.64530.64290.5320.63790.60590.58620.60830.61080.64770.5738
11
ROC0.71450.63990.63030.67610.68110.72490.5782
12
Conf-Matrix
[137 66]
[118 85]
[118 85]
[120 83]
[141 62]
[133 70]
[131 72]
[135 63]
[138 60]
[116 82]
[138 60]
[131 67]
[131 67]
[103 95]
[ 141 57]
[140 58][130 68]
[123 75]
[124 74]
[136 62]
[121 77]
13
[ 74 129]
[ 69 134]
[ 77 126]
[ 77 126]
[ 81 122]
[ 66 137]
[103 100]
[ 78 130]
[ 93 115]
[ 81 127]
[ 88 120]
[ 77 131]
[ 78 130]
[ 95 113]
[ 90 118]
[102 106]
[100 108]
[ 84 124]
[ 84 124]
[ 81 127]
[ 96 112]
14
F10.650.640.610.610.630.670.530.64840.60.60910.620.64530.64190.54320.6160.56980.56250.6090.610.63970.5642
15
16
Accurary Train0.90620.74830.71290.88980.680.69490.57810.99420.96790.9120.99180.84870.86920.66521111111
17
18
19
DOWNSAMPLINGLogR2SVM2RF2
20
3Gramsxxxxxxxx
21
2Gramsxxx
22
embedded2xxx
23
embedded3xxx
24
Doc2Vecxxx
25
Word2Vecxxxxxxxxxxxxx
26
Linguisticxxx
27
Accuracy0.70440.66740.70680.68960.67980.65760.66990.62310.64770.64280.65760.63540.63540.58370.6330.61570.67730.6379
28
ROC0.75410.71180.75570.73230.71860.7004
29
Conf-Matrix
[145 58]
[131 72]
[147 56]
[137 66]
[131 72]
[125 78]
[147 51]
[123 75]
[141 57]
[125 73]
[128 70]
[131 67]
[137 61][115 83][127 71]
[123 75]
[139 59]
[135 63]
30
[ 62 141]
[ 63 140]
[ 63 140]
[ 60 143]
[ 58 145]
[ 61 142]
[ 83 125]
[ 78 130]
[ 86 122]
[ 72 136]
[ 69 139]
[ 81 127]
[ 87 121][ 86 122][ 78 130]
[ 81 127]
[ 72 136]
[ 84 124
31
F10,700,670,700,690,690,670.6510.62950.630.6520.66660.631840.620.59070.6350.6190.67490.6278
32
33
Accurary Train0.82560.77130.71130.680.72450.70970.9920.99580.99170.9950.99750.9958111111
34
35
36
DOWNSAMPLINGLogR3SVM3RF3
37
3Gramsxxxxxxx
38
2Gramsxxxxxxx
39
embedded2xxxxxxx
40
embedded3xxx
41
Doc2Vecxxxxxxx
42
Word2Vecxxxxxxxxxxx
43
Linguisticxxx
44
Accuracy0.65760.68220.68710.67240.67730.6650.62560.68220.69210.63790.67980.66990.60830.60830.6674
45
ROC0.71870.7350.7470.75020.7533
46
Conf-Matrix
[130 73]
[139 64]
[138 65]
[132 71]
[138 65]
[136 62]
[136 62]
[133 65]
[145 53]
[139 59]
[142 56][140 58][119 79]
[123 75]
[143 55]
47
[ 66 137]
[ 65 138]
[ 62 141]
[ 62 141]
[ 66 137]
[ 74 134]
[ 90 118]
[ 64 144]
[ 72 136]
[ 88 120]
[ 74 134][ 76 132][ 80 128]
[ 84 124]
[ 80 128]
48
F10,660,680,690,680.680.6630.6080.69060.6850.620.67330.66330.61680.60930.6547
49
50
Accurary Train0.83140.77050.82480.7590.71870.99340.98840.99670.99750.993411111
51
52
53
DOWNSAMPLINGLogR4SVM4RF4
54
3Gramsxxxxxxxxx
55
2Gramsxxxxxx
56
embedded2xxxxxx
57
embedded3xxxxxx
58
Doc2Vecxxxxxx
59
Word2Vecxxxxxxxxxxxx
60
Linguisticxxx
61
Accuracy0.67980.66740.66250.6650.63790.65760.68470.66
62
ROC0.74680.73160.71820.7362
63
Conf-Matrix
[136 67]
[140 63]
[134 69]
[136 67]
[125 73]
[135 63]
[135 63]
[127 71]
64
[ 63 140]
[ 72 131]
[ 68 135]
[ 69 134]
[ 74 134]
[ 76 132]
[ 65 143]
[ 67 141]
65
F10.680.660.660.660.64570.6550.69080.6714
66
67
Accurary Train0.88730.8560.85030.84450.99670.992590.99670.9934
68
4 feature groups perform worse than best 3 feature groups4 feature groups perform worse than best 3 feature groups
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
Loading...
Main menu