knn ans SVM where svm returns error

Question

nkumar 2013-5-2

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/74369-knn-ans-svm-where-svm-returns-error

评论： Greg Heath 2015-2-10

I am using knn and svm for classification
i get correct answer using knn,but if i use svm it returns wrong result
for examples
my train dataset is 
datsig1 =
          2551        1973
          2816        2887
          7070        6897
          2873        2979
          2588        2675
          1712        2274
           909        4859
           792        4914
          1823        7162
           878        5039
           885        5534
S=
2873        2979
groups=[1;1;1;1;1;1;0;0;0;0;0]
i get 0 if i classify uing svm
kindly help
cl = svmtrain(datsig1,group);
result = svmclassify(cl,S)
result =
       0

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Shashank Prasanna 2013-5-2

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/74369-knn-ans-svm-where-svm-returns-error#answer_84207

在 MATLAB Online 中打开

groups=[1;1;1;1;1;1;0;0;0;0;0]
cl = svmtrain(datsig1,groups);
result = svmclassify(cl,S)
result =
       1

This gave me the right results. What version of MATLAB are you using? Clear your workspace and try it from start.

Also as with all classification algorithm results are completely dependent on your training data and also options in the classification method try different kernels etc.