home bbs files messages ]

Forums before death by AOL, social media and spammers... "We can't have nice things"

   comp.ai      Awaiting the gospel from Sarah Connor      1,954 messages   

[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]

   Message 1,801 of 1,954   
   quasar to All   
   Classification targets question   
   04 Sep 08 10:19:46   
   
   From: anair4@socal.rr.com   
      
   Hi,   
      
     I have a really basic question & the answer is seems fairly obvious   
   to me.  But, I'd like to ask & find out.   
      
     Say, I have a binary training set where I want to detect all 1s.   
   Now, in my training set   
     the frequency of occurrence of 1s is low of the order of 8%.  And,   
   in my test set I   
     have the 1s occurring at about 3% or so.  If the success of detect   
   1s in my training   
     set is ~90%, what should be the success of detecting 1s in my test   
   set approximately.   
      
     Shouldn't it be less than 90%?   I hope my question is clear.   
      
     Thanks   
     Damo.   
      
   [ comp.ai is moderated ... your article may take a while to appear. ]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)   

[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]


(c) 1994,  bbs@darkrealms.ca