昨晚又用SMO重新对上次的训练集做了训练,效果有所改观,结果如下:
Number of instances in the arff file: 9804
Number of attributes: 9302
weka.filters.unsupervised.attribute.ReplaceMissingValues in:9804
weka.filters.unsupervised.attribute.Normalize in:9804
weka.filters.unsupervised.attribute.ReplaceMissingValues in:7843
weka.filters.unsupervised.attribute.Normalize in:7843
weka.filters.unsupervised.attribute.ReplaceMissingValues in:7843
weka.filters.unsupervised.attribute.Normalize in:7843
weka.filters.unsupervised.attribute.ReplaceMissingValues in:7843
weka.filters.unsupervised.attribute.Normalize in:7843
weka.filters.unsupervised.attribute.ReplaceMissingValues in:7843
weka.filters.unsupervised.attribute.Normalize in:7843
weka.filters.unsupervised.attribute.ReplaceMissingValues in:7844
weka.filters.unsupervised.attribute.Normalize in:7844
=== Detailed Accuracy By Class ===
TP Rate FP Rate Precision Recall F-Measure ROC Area Class
0.945 0.004 0.942 0.945 0.944 0.991 C11-Space
0.438 0 0.875 0.438 0.583 0.926 C15-Energy
0.259 0.001 0.538 0.259 0.35 0.946 C16-Electronics
0.64 0 0.889 0.64 0.744 0.98 C17-Communication
0.979 0.01 0.941 0.979 0.959 0.992 C19-Computer
0.394 0 0.929 0.394 0.553 0.975 C23-Mine
0.772 0 0.936 0.772 0.846 0.997 C29-Transport
0.92 0.01 0.881 0.92 0.9 0.992 C3-Art
0.973 0.003 0.975 0.973 0.974 0.994 C31-Enviornment
0.939 0.008 0.931 0.939 0.935 0.988 C32-Agriculture
0.935 0.018 0.912 0.935 0.923 0.983 C34-Economy
0.49 0 0.862 0.49 0.625 0.989 C35-Law
0.725 0 0.925 0.725 0.813 0.986 C36-Medical
0.486 0.001 0.857 0.486 0.621 0.992 C37-Military
0.918 0.018 0.854 0.918 0.885 0.983 C38-Politics
0.943 0.008 0.946 0.943 0.945 0.991 C39-Sports
0.121 0 0.5 0.121 0.195 0.899 C4-Literature
0.407 0.001 0.667 0.407 0.505 0.957 C5-Education
0.591 0 0.929 0.591 0.722 0.882 C6-Philosophy
0.725 0.012 0.749 0.725 0.737 0.965 C7-History
Correctly Classified Instances 8959 91.3811 %
Incorrectly Classified Instances 845 8.6189 %
Kappa statistic 0.9027
Mean absolute error 0.0901
Root mean squared error 0.2085
Relative absolute error 101.4184 %
Root relative squared error 98.9598 %
Total Number of Instances 9804
=== Confusion Matrix ===
a b c d e f g h i j k l m n o p q r s t <-- classified as
605 0 0 0 28 0 0 1 3 1 0 0 0 0 0 1 0 0 0 1 | a = C11-Space
0 14 0 0 3 0 1 0 3 2 6 0 1 0 1 1 0 0 0 0 | b = C15-Energy
1 0 7 1 14 0 1 0 0 0 3 0 0 0 0 0 0 0 0 0 | c = C16-Electronics
0 0 1 16 5 0 0 0 0 1 1 0 0 0 0 0 0 1 0 0 | d = C17-Communication
14 0 5 1 1328 0 0 0 0 0 7 0 0 0 0 2 0 0 0 0 | e = C19-Computer
1 1 0 0 7 13 1 0 1 2 4 0 0 0 2 1 0 0 0 0 | f = C23-Mine
1 1 0 0 1 0 44 0 2 0 2 0 0 0 3 2 0 0 0 1 | g = C29-Transport
0 0 0 0 1 0 0 681 0 0 1 0 0 0 4 13 3 0 0 37 | h = C3-Art
6 0 0 0 2 0 0 0 1184 12 1 0 1 0 8 2 0 0 0 1 | i = C31-Enviornment
2 0 0 0 2 0 0 0 15 959 36 0 1 0 1 2 0 0 0 3 | j = C32-Agriculture
1 0 0 0 3 1 0 0 1 45 1496 3 0 0 28 4 0 1 0 17 | k = C34-Economy
0 0 0 0 1 0 0 0 1 0 5 25 0 1 18 0 0 0 0 0 | l = C35-Law
0 0 0 0 0 0 0 0 1 0 4 0 37 0 5 3 0 1 0 0 | m = C36-Medical
4 0 0 0 0 0 0 0 0 0 1 0 0 36 32 1 0 0 0 0 | n = C37-Military
1 0 0 0 2 0 0 2 1 1 38 0 0 5 940 8 0 1 0 25 | o = C38-Politics
1 0 0 0 6 0 0 9 1 2 16 0 0 0 14 1182 0 3 1 18 | p = C39-Sports
0 0 0 0 2 0 0 14 0 0 1 0 0 0 1 2 4 2 0 7 | q = C4-Literature
0 0 0 0 2 0 0 5 0 0 2 1 0 0 7 16 0 24 1 1 | r = C5-Education
3 0 0 0 2 0 0 3 0 0 0 0 0 0 4 1 0 3 26 2 | s = C6-Philosophy
2 0 0 0 3 0 0 58 1 5 17 0 0 0 33 8 1 0 0 338 | t = C7-History
也贴个截图:
https://wap.sciencenet.cn/blog-713110-571845.html
上一篇:
一个分类模型的评估数据下一篇:
在IDA 5.1中使用“密码算法识别插件 FindCrypt2”