I'm working on a protein classification problem with svm, so i use LibSVM for string data. The string kernel defined into LibSVM is the edit distance kernel, it depends from a parameter gamma. During cross-validation, changing C and gamma parameters, i get 75% of accuracy in every way! Moreover, also changing the number of trainingset patterns, i get the same accuracy. I use the SCOP database. I have no idea about this behavior!
1
votes
Please do not SHOUT your questions here. There are Shift keys on both sides of your keyboard to make them easier to reach, because properly cased text is much easier to read and understand. Typing your subject in ALL CAPS is more difficult to read, will not help you get an answer faster, and quite frankly is rather annoying and rude. Thanks.
– Ken White
I'm so sorry, but i didn't understand what you said me about shift keys. Sorry again.
– Mattia
I mean PLEASE DON'T TYPE YOUR QUESTIONS IN ALL CAPS HERE. :-) The Shift key makes it so you can properly case your text because it's easier to read when it's properly cased. (Use Upper Case where it's appropriate, and don't use it WHERE IT IS NOT.)
– Ken White
1 Answers
0
votes
Look at the counts of each type of misclassification error. If you are getting a constant error rate like this then it is quite possible that every observation is getting assigned to the same class. For example, if 75% of your training observations are in one class and the classifier assigns every observation to that class, then you'll see an error rate of 25%.
